itnroduce dumpless upgrade sprint issue note

5324: Mention utoipa in sprint issues r=curquiza a=irevoire

Update the sprint-issue template to mention the openAPI file and utoipa.

Let me know if something is not clear or missing

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

.github/ISSUE_TEMPLATE/sprint_issue.md

@@ -22,6 +22,28 @@ Related product discussion:
 
 <!---If necessary, create a list with technical/product steps-->
 
+### Reminders when modifying the API
+
+- [ ] Update the openAPI file with utoipa:
+  - [ ] If a new module has been introduced, create a new structure deriving [the OpenAPI proc-macro](https://docs.rs/utoipa/latest/utoipa/derive.OpenApi.html) and nest it in the main [openAPI structure](https://github.com/meilisearch/meilisearch/blob/f2185438eed60fa32d25b15480c5ee064f6fba4a/crates/meilisearch/src/routes/mod.rs#L64-L78).
+  - [ ] If a new route has been introduced, add the [path decorator](https://docs.rs/utoipa/latest/utoipa/attr.path.html) to it and add the route at the top of the file in its openAPI structure.
+  - [ ] If a structure which is deserialized or serialized in the API has been introduced or modified, it must derive the [`schema`](https://docs.rs/utoipa/latest/utoipa/macro.schema.html) or the [`IntoParams`](https://docs.rs/utoipa/latest/utoipa/derive.IntoParams.html) proc-macro.
+        If it's a **new** structure you must also add it to the big list of structures [in the main `OpenApi` structure](https://github.com/meilisearch/meilisearch/blob/f2185438eed60fa32d25b15480c5ee064f6fba4a/crates/meilisearch/src/routes/mod.rs#L88).
+  - [ ] Once everything is done, start Meilisearch with the swagger flag: `cargo run --features swagger`, open `http://localhost:7700/scalar` on your browser, and ensure everything works as expected.
+  - For more info, refer to [this presentation](https://pitch.com/v/generating-the-openapi-file-jrn3nh).
+
+### Reminders when modifying a database
+
+- [ ] If modifying the API keys => Reach out to the team we've not planned anything for that yet
+- [ ] If modifying a milli index =>
+  - [ ] You can add the upgrade code to go from the previous to the latest version [here](https://github.com/meilisearch/meilisearch/blob/main/crates/milli/src/update/upgrade/mod.rs)
+  - [ ] Don't forget to return `true` to update the stats if needed
+  - [ ] Did the read read part requires a change? (gettings documents, search, facet search etc)
+        The index must **always** be able to read the old version of the database. Make everything with that in mind
+        and if the old and new database are not compatible you may need to match on the version of the database and keep both the old and new code. (The version of the index is stored in the main database under the `VERSION_KEY` key)
+- [ ] If modifying the task queue or the index-mapper =>
+  - [ ] Update 
+
 ### Reminders when modifying the Setting API
 
 <!--- Special steps to remind when adding a new index setting -->

.github/workflows/test-suite.yml

@@ -4,7 +4,7 @@ on:
   workflow_dispatch:
   schedule:
     # Everyday at 5:00am
-    - cron: '0 5 * * *'
+    - cron: "0 5 * * *"
   pull_request:
   push:
     # trying and staging branches are for Bors config
@@ -84,10 +84,42 @@ jobs:
       - uses: dtolnay/rust-toolchain@1.81
       - name: Run cargo build with almost all features
         run: |
-          cargo build --workspace --locked --release --features "$(cargo xtask list-features --exclude-feature cuda)"
+          cargo build --workspace --locked --release --features "$(cargo xtask list-features --exclude-feature cuda,test-ollama)"
       - name: Run cargo test with almost all features
         run: |
-          cargo test --workspace --locked --release --features "$(cargo xtask list-features --exclude-feature cuda)"
+          cargo test --workspace --locked --release --features "$(cargo xtask list-features --exclude-feature cuda,test-ollama)"
+
+  ollama-ubuntu:
+    name: Test with Ollama
+    runs-on: ubuntu-latest
+    env:
+      MEILI_TEST_OLLAMA_SERVER: "http://localhost:11434"
+    steps:
+      - uses: actions/checkout@v1
+      - name: Install Ollama
+        run: |
+          curl -fsSL https://ollama.com/install.sh | sudo -E sh
+      - name: Start serving
+        run: |
+          # Run it in the background, there is no way to daemonise at the moment
+          ollama serve &
+
+          # A short pause is required before the HTTP port is opened
+          sleep 5
+
+          # This endpoint blocks until ready
+          time curl -i http://localhost:11434
+
+      - name: Pull nomic-embed-text & all-minilm
+        run: |
+          ollama pull nomic-embed-text
+          ollama pull all-minilm
+
+      - name: Run cargo test
+        uses: actions-rs/cargo@v1
+        with:
+          command: test
+          args: --locked --release --all --features test-ollama ollama
 
   test-disabled-tokenization:
     name: Test disabled tokenization

crates/benchmarks/benches/indexing.rs

@@ -10,7 +10,7 @@ use milli::documents::PrimaryKey;
 use milli::heed::{EnvOpenOptions, RwTxn};
 use milli::progress::Progress;
 use milli::update::new::indexer;
-use milli::update::{IndexDocumentsMethod, IndexerConfig, Settings};
+use milli::update::{IndexerConfig, Settings};
 use milli::vector::EmbeddingConfigs;
 use milli::Index;
 use rand::seq::SliceRandom;
@@ -138,10 +138,9 @@ fn indexing_songs_default(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_SONGS, "csv");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -205,10 +204,9 @@ fn reindexing_songs_default(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_SONGS, "csv");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -250,10 +248,9 @@ fn reindexing_songs_default(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_SONGS, "csv");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -319,10 +316,9 @@ fn deleting_songs_in_batches_default(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_SONGS, "csv");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -396,10 +392,9 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_SONGS_1_2, "csv");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -441,10 +436,9 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_SONGS_3_4, "csv");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -482,10 +476,9 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_SONGS_4_4, "csv");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -549,11 +542,10 @@ fn indexing_songs_without_faceted_numbers(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_SONGS, "csv");
 
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -617,10 +609,9 @@ fn indexing_songs_without_faceted_fields(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_SONGS, "csv");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -684,10 +675,9 @@ fn indexing_wiki(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_WIKI_ARTICLES, "csv");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -750,10 +740,9 @@ fn reindexing_wiki(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_WIKI_ARTICLES, "csv");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -795,10 +784,9 @@ fn reindexing_wiki(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_WIKI_ARTICLES, "csv");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -863,10 +851,9 @@ fn deleting_wiki_in_batches_default(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_WIKI_ARTICLES, "csv");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -939,11 +926,10 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents =
                     utils::documents_from(datasets_paths::SMOL_WIKI_ARTICLES_1_2, "csv");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -985,11 +971,10 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents =
                     utils::documents_from(datasets_paths::SMOL_WIKI_ARTICLES_3_4, "csv");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -1027,11 +1012,10 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents =
                     utils::documents_from(datasets_paths::SMOL_WIKI_ARTICLES_4_4, "csv");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -1095,10 +1079,9 @@ fn indexing_movies_default(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::MOVIES, "json");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -1161,10 +1144,9 @@ fn reindexing_movies_default(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::MOVIES, "json");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -1206,10 +1188,9 @@ fn reindexing_movies_default(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::MOVIES, "json");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -1274,10 +1255,9 @@ fn deleting_movies_in_batches_default(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::MOVIES, "json");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -1387,10 +1367,9 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::MOVIES_1_2, "json");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -1432,10 +1411,9 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::MOVIES_3_4, "json");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -1473,10 +1451,9 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::MOVIES_4_4, "json");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -1563,10 +1540,9 @@ fn indexing_nested_movies_default(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::NESTED_MOVIES, "json");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -1654,10 +1630,9 @@ fn deleting_nested_movies_in_batches_default(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::NESTED_MOVIES, "json");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -1737,10 +1712,9 @@ fn indexing_nested_movies_without_faceted_fields(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::NESTED_MOVIES, "json");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -1804,10 +1778,9 @@ fn indexing_geo(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_ALL_COUNTRIES, "jsonl");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -1870,10 +1843,9 @@ fn reindexing_geo(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_ALL_COUNTRIES, "jsonl");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -1915,10 +1887,9 @@ fn reindexing_geo(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_ALL_COUNTRIES, "jsonl");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer
@@ -1983,10 +1954,9 @@ fn deleting_geo_in_batches_default(c: &mut Criterion) {
                 let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-                let mut indexer =
-                    indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+                let mut indexer = indexer::DocumentOperation::new();
                 let documents = utils::documents_from(datasets_paths::SMOL_ALL_COUNTRIES, "jsonl");
-                indexer.add_documents(&documents).unwrap();
+                indexer.replace_documents(&documents).unwrap();
 
                 let indexer_alloc = Bump::new();
                 let (document_changes, _operation_stats, primary_key) = indexer

crates/benchmarks/benches/utils.rs

@@ -12,7 +12,7 @@ use memmap2::Mmap;
 use milli::heed::EnvOpenOptions;
 use milli::progress::Progress;
 use milli::update::new::indexer;
-use milli::update::{IndexDocumentsMethod, IndexerConfig, Settings};
+use milli::update::{IndexerConfig, Settings};
 use milli::vector::EmbeddingConfigs;
 use milli::{Criterion, Filter, Index, Object, TermsMatchingStrategy};
 use serde_json::Value;
@@ -99,8 +99,8 @@ pub fn base_setup(conf: &Conf) -> Index {
     let mut new_fields_ids_map = db_fields_ids_map.clone();
 
     let documents = documents_from(conf.dataset, conf.dataset_format);
-    let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
-    indexer.add_documents(&documents).unwrap();
+    let mut indexer = indexer::DocumentOperation::new();
+    indexer.replace_documents(&documents).unwrap();
 
     let indexer_alloc = Bump::new();
     let (document_changes, _operation_stats, primary_key) = indexer

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

@@ -12,7 +12,7 @@ use milli::documents::mmap_from_objects;
 use milli::heed::EnvOpenOptions;
 use milli::progress::Progress;
 use milli::update::new::indexer;
-use milli::update::{IndexDocumentsMethod, IndexerConfig};
+use milli::update::IndexerConfig;
 use milli::vector::EmbeddingConfigs;
 use milli::Index;
 use serde_json::Value;
@@ -89,9 +89,7 @@ fn main() {
 
                             let indexer_alloc = Bump::new();
                             let embedders = EmbeddingConfigs::default();
-                            let mut indexer = indexer::DocumentOperation::new(
-                                IndexDocumentsMethod::ReplaceDocuments,
-                            );
+                            let mut indexer = indexer::DocumentOperation::new();
 
                             let mut operations = Vec::new();
                             for op in batch.0 {
@@ -115,7 +113,7 @@ fn main() {
                             for op in &operations {
                                 match op {
                                     Either::Left(documents) => {
-                                        indexer.add_documents(documents).unwrap()
+                                        indexer.replace_documents(documents).unwrap()
                                     }
                                     Either::Right(ids) => indexer.delete_documents(ids),
                                 }

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

@@ -5,12 +5,8 @@ tasks affecting a single index into a [batch](crate::batch::Batch).
 The main function of the autobatcher is [`next_autobatch`].
 */
 
-use std::ops::ControlFlow::{self, Break, Continue};
-
-use meilisearch_types::milli::update::IndexDocumentsMethod::{
-    self, ReplaceDocuments, UpdateDocuments,
-};
 use meilisearch_types::tasks::TaskId;
+use std::ops::ControlFlow::{self, Break, Continue};
 
 use crate::KindWithContent;
 
@@ -19,19 +15,11 @@ use crate::KindWithContent;
 ///
 /// Only the non-prioritised tasks that can be grouped in a batch have a corresponding [`AutobatchKind`]
 enum AutobatchKind {
-    DocumentImport {
-        method: IndexDocumentsMethod,
-        allow_index_creation: bool,
-        primary_key: Option<String>,
-    },
+    DocumentImport { allow_index_creation: bool, primary_key: Option<String> },
     DocumentEdition,
-    DocumentDeletion {
-        by_filter: bool,
-    },
+    DocumentDeletion { by_filter: bool },
     DocumentClear,
-    Settings {
-        allow_index_creation: bool,
-    },
+    Settings { allow_index_creation: bool },
     IndexCreation,
     IndexDeletion,
     IndexUpdate,
@@ -60,11 +48,8 @@ impl From<KindWithContent> for AutobatchKind {
     fn from(kind: KindWithContent) -> Self {
         match kind {
             KindWithContent::DocumentAdditionOrUpdate {
-                method,
-                allow_index_creation,
-                primary_key,
-                ..
-            } => AutobatchKind::DocumentImport { method, allow_index_creation, primary_key },
+                allow_index_creation, primary_key, ..
+            } => AutobatchKind::DocumentImport { allow_index_creation, primary_key },
             KindWithContent::DocumentEdition { .. } => AutobatchKind::DocumentEdition,
             KindWithContent::DocumentDeletion { .. } => {
                 AutobatchKind::DocumentDeletion { by_filter: false }
@@ -99,7 +84,6 @@ pub enum BatchKind {
         ids: Vec<TaskId>,
     },
     DocumentOperation {
-        method: IndexDocumentsMethod,
         allow_index_creation: bool,
         primary_key: Option<String>,
         operation_ids: Vec<TaskId>,
@@ -172,12 +156,11 @@ impl BatchKind {
             K::IndexUpdate => (Break(BatchKind::IndexUpdate { id: task_id }), false),
             K::IndexSwap => (Break(BatchKind::IndexSwap { id: task_id }), false),
             K::DocumentClear => (Continue(BatchKind::DocumentClear { ids: vec![task_id] }), false),
-            K::DocumentImport { method, allow_index_creation, primary_key: pk }
+            K::DocumentImport { allow_index_creation, primary_key: pk }
                 if primary_key.is_none() || pk.is_none() || primary_key == pk.as_deref() =>
             {
                 (
                     Continue(BatchKind::DocumentOperation {
-                        method,
                         allow_index_creation,
                         primary_key: pk,
                         operation_ids: vec![task_id],
@@ -186,9 +169,8 @@ impl BatchKind {
                 )
             }
             // if the primary key set in the task was different than ours we should stop and make this batch fail asap.
-            K::DocumentImport { method, allow_index_creation, primary_key } => (
+            K::DocumentImport { allow_index_creation, primary_key } => (
                 Break(BatchKind::DocumentOperation {
-                    method,
                     allow_index_creation,
                     primary_key,
                     operation_ids: vec![task_id],
@@ -257,7 +239,7 @@ impl BatchKind {
             (
                 BatchKind::DocumentClear { mut ids }
                 | BatchKind::DocumentDeletion { deletion_ids: mut ids, includes_by_filter: _ }
-                | BatchKind::DocumentOperation { method: _, allow_index_creation: _, primary_key: _, operation_ids: mut ids }
+                | BatchKind::DocumentOperation { allow_index_creation: _, primary_key: _, operation_ids: mut ids }
                 | BatchKind::Settings { allow_index_creation: _, settings_ids: mut ids },
                 K::IndexDeletion,
             ) => {
@@ -285,46 +267,32 @@ impl BatchKind {
                 K::DocumentImport { .. } | K::Settings { .. },
             ) => Break(this),
             (
-                BatchKind::DocumentOperation { method: _, allow_index_creation: _, primary_key: _, mut operation_ids },
+                BatchKind::DocumentOperation { allow_index_creation: _, primary_key: _, mut operation_ids },
                 K::DocumentClear,
             ) => {
                 operation_ids.push(id);
                 Continue(BatchKind::DocumentClear { ids: operation_ids })
             }
 
-            // we can autobatch the same kind of document additions / updates
+            // we can autobatch different kind of document operations and mix replacements with updates
             (
-                BatchKind::DocumentOperation { method: ReplaceDocuments, allow_index_creation, primary_key: _, mut operation_ids },
-                K::DocumentImport { method: ReplaceDocuments, primary_key: pk, .. },
+                BatchKind::DocumentOperation { allow_index_creation, primary_key: _, mut operation_ids },
+                K::DocumentImport { primary_key: pk, .. },
             ) => {
                 operation_ids.push(id);
                 Continue(BatchKind::DocumentOperation {
-                    method: ReplaceDocuments,
                     allow_index_creation,
                     operation_ids,
                     primary_key: pk,
                 })
             }
             (
-                BatchKind::DocumentOperation { method: UpdateDocuments, allow_index_creation, primary_key: _, mut operation_ids },
-                K::DocumentImport { method: UpdateDocuments, primary_key: pk, .. },
-            ) => {
-                operation_ids.push(id);
-                Continue(BatchKind::DocumentOperation {
-                    method: UpdateDocuments,
-                    allow_index_creation,
-                    primary_key: pk,
-                    operation_ids,
-                })
-            }
-            (
-                BatchKind::DocumentOperation { method, allow_index_creation, primary_key, mut operation_ids },
+                BatchKind::DocumentOperation { allow_index_creation, primary_key, mut operation_ids },
                 K::DocumentDeletion { by_filter: false },
             ) => {
                 operation_ids.push(id);
 
                 Continue(BatchKind::DocumentOperation {
-                    method,
                     allow_index_creation,
                     primary_key,
                     operation_ids,
@@ -337,13 +305,6 @@ impl BatchKind {
             ) => {
                 Break(this)
             }
-            // but we can't autobatch documents if it's not the same kind
-            // this match branch MUST be AFTER the previous one
-            (
-                this @ BatchKind::DocumentOperation { .. },
-                K::DocumentImport { .. },
-            ) => Break(this),
-
             (
                 this @ BatchKind::DocumentOperation { .. },
                 K::Settings { .. },
@@ -361,12 +322,11 @@ impl BatchKind {
             // we can autobatch the deletion and import if the index already exists
             (
                 BatchKind::DocumentDeletion { mut deletion_ids, includes_by_filter: false },
-                K::DocumentImport { method, allow_index_creation, primary_key }
+                K::DocumentImport { allow_index_creation, primary_key }
             ) if index_already_exists => {
                 deletion_ids.push(id);
 
                 Continue(BatchKind::DocumentOperation {
-                    method,
                     allow_index_creation,
                     primary_key,
                     operation_ids: deletion_ids,
@@ -375,12 +335,11 @@ impl BatchKind {
             // we can autobatch the deletion and import if both can't create an index
             (
                 BatchKind::DocumentDeletion { mut deletion_ids, includes_by_filter: false },
-                K::DocumentImport { method, allow_index_creation, primary_key }
+                K::DocumentImport { allow_index_creation, primary_key }
             ) if !allow_index_creation => {
                 deletion_ids.push(id);
 
                 Continue(BatchKind::DocumentOperation {
-                    method,
                     allow_index_creation,
                     primary_key,
                     operation_ids: deletion_ids,

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

@@ -92,29 +92,29 @@ fn idx_swap() -> KindWithContent {
 fn autobatch_simple_operation_together() {
     // we can autobatch one or multiple `ReplaceDocuments` together.
     // if the index exists.
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_imp( ReplaceDocuments, true , None), doc_imp(ReplaceDocuments, true , None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1, 2] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), doc_imp( ReplaceDocuments, false , None), doc_imp(ReplaceDocuments, false , None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1, 2] }, false))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_imp( ReplaceDocuments, true , None), doc_imp(ReplaceDocuments, true , None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1, 2] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), doc_imp( ReplaceDocuments, false , None), doc_imp(ReplaceDocuments, false , None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0, 1, 2] }, false))");
 
     // if it doesn't exists.
-    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
-    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, true, None), doc_imp( ReplaceDocuments, true , None), doc_imp(ReplaceDocuments, true , None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1, 2] }, true))");
-    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, false, None), doc_imp( ReplaceDocuments, true , None), doc_imp(ReplaceDocuments, true , None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, true, None), doc_imp( ReplaceDocuments, true , None), doc_imp(ReplaceDocuments, true , None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1, 2] }, true))");
+    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, false, None), doc_imp( ReplaceDocuments, true , None), doc_imp(ReplaceDocuments, true , None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
 
     // we can autobatch one or multiple `UpdateDocuments` together.
     // if the index exists.
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), doc_imp(UpdateDocuments, true, None), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1, 2] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None), doc_imp(UpdateDocuments, false, None), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1, 2] }, false))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), doc_imp(UpdateDocuments, true, None), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1, 2] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None), doc_imp(UpdateDocuments, false, None), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0, 1, 2] }, false))");
 
     // if it doesn't exists.
-    debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, true, None), doc_imp(UpdateDocuments, true, None), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1, 2] }, true))");
-    debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
-    debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, false, None), doc_imp(UpdateDocuments, false, None), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1, 2] }, false))");
+    debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, true, None), doc_imp(UpdateDocuments, true, None), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1, 2] }, true))");
+    debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+    debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, false, None), doc_imp(UpdateDocuments, false, None), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0, 1, 2] }, false))");
 
     // we can autobatch one or multiple DocumentDeletion together
     debug_snapshot!(autobatch_from(true, None, [doc_del()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: false }, false))");
@@ -140,53 +140,53 @@ fn autobatch_simple_operation_together() {
     debug_snapshot!(autobatch_from(false,None,  [settings(false), settings(false), settings(false)]), @"Some((Settings { allow_index_creation: false, settings_ids: [0, 1, 2] }, false))");
 
     // We can autobatch document addition with document deletion
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_del()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), doc_del()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), doc_del()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None), doc_del()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, true))"###);
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, true))"###);
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, true, None), doc_del()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, true, None), doc_del()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, false, None), doc_del()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, false, None), doc_del()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, true))"###);
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, true))"###);
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_del()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), doc_del()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), doc_del()]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None), doc_del()]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, true))"###);
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, true))"###);
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, true, None), doc_del()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, true, None), doc_del()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, false, None), doc_del()]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, false, None), doc_del()]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, true))"###);
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, true))"###);
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
     // And the other way around
-    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, false))");
-    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, false))");
-    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
-    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
-    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, true, Some("catto"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
-    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, true, Some("catto"))]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
-    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, false, Some("catto"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
-    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, false, Some("catto"))]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
-    debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
-    debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
-    debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(ReplaceDocuments, false, Some("catto"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
-    debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(UpdateDocuments, false, Some("catto"))]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, false))");
+    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, false))");
+    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, true, Some("catto"))]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, true, Some("catto"))]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, false, Some("catto"))]), @r###"Some((DocumentOperation { allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+    debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, false, Some("catto"))]), @r###"Some((DocumentOperation { allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+    debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+    debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+    debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(ReplaceDocuments, false, Some("catto"))]), @r###"Some((DocumentOperation { allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+    debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(UpdateDocuments, false, Some("catto"))]), @r###"Some((DocumentOperation { allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
 
     // But we can't autobatch document addition with document deletion by filter
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), doc_del_fil()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None), doc_del_fil()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, false, None), doc_del_fil()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, false, None), doc_del_fil()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, true, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, true, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, false, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
-    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, false, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), doc_del_fil()]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None), doc_del_fil()]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, false, None), doc_del_fil()]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, false, None), doc_del_fil()]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, true, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, true, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, false, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
+    debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, false, Some("catto")), doc_del_fil()]), @r###"Some((DocumentOperation { allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
     // And the other way around
     debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
     debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
@@ -203,27 +203,27 @@ fn autobatch_simple_operation_together() {
 }
 
 #[test]
-fn simple_document_operation_dont_autobatch_with_other() {
-    // addition, updates and deletion by filter can't batch together
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+fn simple_different_document_operations_autobatch_together() {
+    // addition and updates with deletion by filter can't batch together
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_del_fil()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
     debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
     debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
 
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), idx_create()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), idx_create()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), idx_create()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), idx_create()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
     debug_snapshot!(autobatch_from(true, None, [doc_del(), idx_create()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: false }, false))");
     debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), idx_create()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
 
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), idx_update()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), idx_update()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), idx_update()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), idx_update()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
     debug_snapshot!(autobatch_from(true, None, [doc_del(), idx_update()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: false }, false))");
     debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), idx_update()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
 
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), idx_swap()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), idx_swap()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), idx_swap()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), idx_swap()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
     debug_snapshot!(autobatch_from(true, None, [doc_del(), idx_swap()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: false }, false))");
     debug_snapshot!(autobatch_from(true, None, [doc_del_fil(), idx_swap()]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: true }, false))");
 }
@@ -231,28 +231,28 @@ fn simple_document_operation_dont_autobatch_with_other() {
 #[test]
 fn document_addition_doesnt_batch_with_settings() {
     // simple case
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
 
     // multiple settings and doc addition
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, None), settings(true), settings(true)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, None), settings(true), settings(true)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, None), settings(true), settings(true)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, None), settings(true), settings(true)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
 
     // addition and setting unordered
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), doc_imp(ReplaceDocuments, true, None), settings(true)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), doc_imp(UpdateDocuments, true, None), settings(true)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), doc_imp(ReplaceDocuments, true, None), settings(true)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), doc_imp(UpdateDocuments, true, None), settings(true)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
 
     // Doesn't batch with other forbidden operations
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), doc_del()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), doc_del()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), idx_create()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), idx_create()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), idx_update()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), idx_update()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), idx_swap()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), idx_swap()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), doc_del()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), doc_del()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), idx_create()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), idx_create()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), idx_update()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), idx_update()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), idx_swap()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), idx_swap()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
 }
 
 #[test]
@@ -280,8 +280,8 @@ fn clear_and_additions_and_settings() {
     debug_snapshot!(autobatch_from(true, None, [doc_clr(), settings(true)]), @"Some((DocumentClear { ids: [0] }, false))");
 
     debug_snapshot!(autobatch_from(true, None, [settings(true), doc_clr(), settings(true)]), @"Some((ClearAndSettings { other: [1], allow_index_creation: true, settings_ids: [0, 2] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), doc_clr()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), doc_clr()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), doc_clr()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), doc_clr()]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
 }
 
 #[test]
@@ -333,17 +333,17 @@ fn anything_and_index_deletion() {
 #[test]
 fn allowed_and_disallowed_index_creation() {
     // `DocumentImport` can't be mixed with those disallowed to do so except if the index already exists.
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), settings(true)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), settings(true)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
 
-    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, false, None), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
-    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
-    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, false, None), doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
-    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, true, None), settings(true)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, false, None), settings(true)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, false, None), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
+    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, false, None), doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, true, None), settings(true)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, false, None), settings(true)]), @"Some((DocumentOperation { allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
 
     // batch deletion and addition
     debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(ReplaceDocuments, true, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0], includes_by_filter: false }, false))");
@@ -356,40 +356,40 @@ fn allowed_and_disallowed_index_creation() {
 fn autobatch_primary_key() {
     // ==> If I have a pk
     // With a single update
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("id"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("id"), operation_ids: [0] }, true))"###);
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("other"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("id"))]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("id"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("other"))]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
 
     // With a multiple updates
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, Some("id"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("id"), operation_ids: [0, 1] }, true))"###);
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("id"), operation_ids: [0, 1] }, true))"###);
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, Some("other"))]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("id"))]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, Some("id"))]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("id"), operation_ids: [0, 1] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("id"), operation_ids: [0, 1] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, Some("other"))]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("id"))]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
 
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("id"), operation_ids: [0] }, true))"###);
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, Some("id"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("id"), operation_ids: [0, 1] }, true))"###);
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("id"), operation_ids: [0, 1] }, true))"###);
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, Some("other"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("id"), operation_ids: [0] }, true))"###);
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("id"), operation_ids: [0] }, true))"###);
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("id"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("id"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("id"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, Some("id"))]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("id"), operation_ids: [0, 1] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("id"), operation_ids: [0, 1] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, Some("other"))]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("id"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("id"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("id"))]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("id"), operation_ids: [0] }, true))"###);
 
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("id"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("other"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
-    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("id"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("id"))]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("other"))]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, Some("id"), [doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("other")), doc_imp(ReplaceDocuments, true, Some("id"))]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
 
     // ==> If I don't have a pk
     // With a single update
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, Some("id"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("id"), operation_ids: [0] }, true))"###);
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, Some("other"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, Some("id"))]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("id"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, Some("other"))]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("other"), operation_ids: [0] }, true))"###);
 
     // With a multiple updates
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, Some("id"))]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("id"), operation_ids: [0] }, true))"###);
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_imp(ReplaceDocuments, true, Some("id"))]), @"Some((DocumentOperation { allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+    debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, Some("id")), doc_imp(ReplaceDocuments, true, None)]), @r###"Some((DocumentOperation { allow_index_creation: true, primary_key: Some("id"), operation_ids: [0] }, true))"###);
 }

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

@@ -54,7 +54,8 @@ pub(crate) enum Batch {
 
 #[derive(Debug)]
 pub(crate) enum DocumentOperation {
-    Add(Uuid),
+    Replace(Uuid),
+    Update(Uuid),
     Delete(Vec<String>),
 }
 
@@ -64,7 +65,6 @@ pub(crate) enum IndexOperation {
     DocumentOperation {
         index_uid: String,
         primary_key: Option<String>,
-        method: IndexDocumentsMethod,
         operations: Vec<DocumentOperation>,
         tasks: Vec<Task>,
     },
@@ -254,7 +254,7 @@ impl IndexScheduler {
                     _ => unreachable!(),
                 }
             }
-            BatchKind::DocumentOperation { method, operation_ids, .. } => {
+            BatchKind::DocumentOperation { operation_ids, .. } => {
                 let tasks = self.queue.get_existing_tasks_for_processing_batch(
                     rtxn,
                     current_batch,
@@ -276,9 +276,17 @@ impl IndexScheduler {
 
                 for task in tasks.iter() {
                     match task.kind {
-                        KindWithContent::DocumentAdditionOrUpdate { content_file, .. } => {
-                            operations.push(DocumentOperation::Add(content_file));
-                        }
+                        KindWithContent::DocumentAdditionOrUpdate {
+                            content_file, method, ..
+                        } => match method {
+                            IndexDocumentsMethod::ReplaceDocuments => {
+                                operations.push(DocumentOperation::Replace(content_file))
+                            }
+                            IndexDocumentsMethod::UpdateDocuments => {
+                                operations.push(DocumentOperation::Update(content_file))
+                            }
+                            _ => unreachable!("Unknown document merging method"),
+                        },
                         KindWithContent::DocumentDeletion { ref documents_ids, .. } => {
                             operations.push(DocumentOperation::Delete(documents_ids.clone()));
                         }
@@ -290,7 +298,6 @@ impl IndexScheduler {
                     op: IndexOperation::DocumentOperation {
                         index_uid,
                         primary_key,
-                        method,
                         operations,
                         tasks,
                     },

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

@@ -62,23 +62,21 @@ impl IndexScheduler {
 
                 Ok(tasks)
             }
-            IndexOperation::DocumentOperation {
-                index_uid,
-                primary_key,
-                method,
-                operations,
-                mut tasks,
-            } => {
+            IndexOperation::DocumentOperation { index_uid, primary_key, operations, mut tasks } => {
                 progress.update_progress(DocumentOperationProgress::RetrievingConfig);
                 // TODO: at some point, for better efficiency we might want to reuse the bumpalo for successive batches.
                 // this is made difficult by the fact we're doing private clones of the index scheduler and sending it
                 // to a fresh thread.
                 let mut content_files = Vec::new();
                 for operation in &operations {
-                    if let DocumentOperation::Add(content_uuid) = operation {
-                        let content_file = self.queue.file_store.get_update(*content_uuid)?;
-                        let mmap = unsafe { memmap2::Mmap::map(&content_file)? };
-                        content_files.push(mmap);
+                    match operation {
+                        DocumentOperation::Replace(content_uuid)
+                        | DocumentOperation::Update(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);
+                        }
+                        _ => (),
                     }
                 }
 
@@ -87,17 +85,23 @@ impl IndexScheduler {
                 let mut new_fields_ids_map = db_fields_ids_map.clone();
 
                 let mut content_files_iter = content_files.iter();
-                let mut indexer = indexer::DocumentOperation::new(method);
+                let mut indexer = indexer::DocumentOperation::new();
                 let embedders = index
                     .embedding_configs(index_wtxn)
                     .map_err(|e| Error::from_milli(e, Some(index_uid.clone())))?;
                 let embedders = self.embedders(index_uid.clone(), embedders)?;
                 for operation in operations {
                     match operation {
-                        DocumentOperation::Add(_content_uuid) => {
+                        DocumentOperation::Replace(_content_uuid) => {
                             let mmap = content_files_iter.next().unwrap();
                             indexer
-                                .add_documents(mmap)
+                                .replace_documents(mmap)
+                                .map_err(|e| Error::from_milli(e, Some(index_uid.clone())))?;
+                        }
+                        DocumentOperation::Update(_content_uuid) => {
+                            let mmap = content_files_iter.next().unwrap();
+                            indexer
+                                .update_documents(mmap)
                                 .map_err(|e| Error::from_milli(e, Some(index_uid.clone())))?;
                         }
                         DocumentOperation::Delete(document_ids) => {

crates/index-scheduler/src/scheduler/snapshots/test_document_addition.rs/test_mixed_document_addition/all_tasks_processed.snap

@@ -1,6 +1,5 @@
 ---
 source: crates/index-scheduler/src/scheduler/test_document_addition.rs
-snapshot_kind: text
 ---
 ### Autobatching Enabled = true
 ### Processing batch None:
@@ -8,15 +7,15 @@ snapshot_kind: text
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, batch_uid: 0, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}
-1 {uid: 1, batch_uid: 1, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000001, documents_count: 1, allow_index_creation: true }}
-2 {uid: 2, batch_uid: 2, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000002, documents_count: 1, allow_index_creation: true }}
-3 {uid: 3, batch_uid: 3, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000003, documents_count: 1, allow_index_creation: true }}
-4 {uid: 4, batch_uid: 4, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000004, documents_count: 1, allow_index_creation: true }}
-5 {uid: 5, batch_uid: 5, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000005, documents_count: 1, allow_index_creation: true }}
-6 {uid: 6, batch_uid: 6, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000006, documents_count: 1, allow_index_creation: true }}
-7 {uid: 7, batch_uid: 7, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000007, documents_count: 1, allow_index_creation: true }}
-8 {uid: 8, batch_uid: 8, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000008, documents_count: 1, allow_index_creation: true }}
-9 {uid: 9, batch_uid: 9, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000009, documents_count: 1, allow_index_creation: true }}
+1 {uid: 1, batch_uid: 0, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000001, documents_count: 1, allow_index_creation: true }}
+2 {uid: 2, batch_uid: 0, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000002, documents_count: 1, allow_index_creation: true }}
+3 {uid: 3, batch_uid: 0, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000003, documents_count: 1, allow_index_creation: true }}
+4 {uid: 4, batch_uid: 0, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000004, documents_count: 1, allow_index_creation: true }}
+5 {uid: 5, batch_uid: 0, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000005, documents_count: 1, allow_index_creation: true }}
+6 {uid: 6, batch_uid: 0, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000006, documents_count: 1, allow_index_creation: true }}
+7 {uid: 7, batch_uid: 0, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000007, documents_count: 1, allow_index_creation: true }}
+8 {uid: 8, batch_uid: 0, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: UpdateDocuments, content_file: 00000000-0000-0000-0000-000000000008, documents_count: 1, allow_index_creation: true }}
+9 {uid: 9, batch_uid: 0, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000009, documents_count: 1, allow_index_creation: true }}
 ----------------------------------------------------------------------
 ### Status:
 enqueued []
@@ -48,97 +47,35 @@ doggos: { number_of_documents: 10, field_distribution: {"doggo": 10, "id": 10} }
 [timestamp] [9,]
 ----------------------------------------------------------------------
 ### Started At:
-[timestamp] [0,]
-[timestamp] [1,]
-[timestamp] [2,]
-[timestamp] [3,]
-[timestamp] [4,]
-[timestamp] [5,]
-[timestamp] [6,]
-[timestamp] [7,]
-[timestamp] [8,]
-[timestamp] [9,]
+[timestamp] [0,1,2,3,4,5,6,7,8,9,]
 ----------------------------------------------------------------------
 ### Finished At:
-[timestamp] [0,]
-[timestamp] [1,]
-[timestamp] [2,]
-[timestamp] [3,]
-[timestamp] [4,]
-[timestamp] [5,]
-[timestamp] [6,]
-[timestamp] [7,]
-[timestamp] [8,]
-[timestamp] [9,]
+[timestamp] [0,1,2,3,4,5,6,7,8,9,]
 ----------------------------------------------------------------------
 ### All Batches:
-0 {uid: 0, details: {"receivedDocuments":1,"indexedDocuments":1}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
-1 {uid: 1, details: {"receivedDocuments":1,"indexedDocuments":1}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
-2 {uid: 2, details: {"receivedDocuments":1,"indexedDocuments":1}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
-3 {uid: 3, details: {"receivedDocuments":1,"indexedDocuments":1}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
-4 {uid: 4, details: {"receivedDocuments":1,"indexedDocuments":1}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
-5 {uid: 5, details: {"receivedDocuments":1,"indexedDocuments":1}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
-6 {uid: 6, details: {"receivedDocuments":1,"indexedDocuments":1}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
-7 {uid: 7, details: {"receivedDocuments":1,"indexedDocuments":1}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
-8 {uid: 8, details: {"receivedDocuments":1,"indexedDocuments":1}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
-9 {uid: 9, details: {"receivedDocuments":1,"indexedDocuments":1}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
+0 {uid: 0, details: {"receivedDocuments":10,"indexedDocuments":10}, stats: {"totalNbTasks":10,"status":{"succeeded":10},"types":{"documentAdditionOrUpdate":10},"indexUids":{"doggos":10}}, }
 ----------------------------------------------------------------------
 ### Batch to tasks mapping:
-0 [0,]
-1 [1,]
-2 [2,]
-3 [3,]
-4 [4,]
-5 [5,]
-6 [6,]
-7 [7,]
-8 [8,]
-9 [9,]
+0 [0,1,2,3,4,5,6,7,8,9,]
 ----------------------------------------------------------------------
 ### Batches Status:
-succeeded [0,1,2,3,4,5,6,7,8,9,]
+succeeded [0,]
 ----------------------------------------------------------------------
 ### Batches Kind:
-"documentAdditionOrUpdate" [0,1,2,3,4,5,6,7,8,9,]
+"documentAdditionOrUpdate" [0,]
 ----------------------------------------------------------------------
 ### Batches Index Tasks:
-doggos [0,1,2,3,4,5,6,7,8,9,]
+doggos [0,]
 ----------------------------------------------------------------------
 ### Batches Enqueued At:
 [timestamp] [0,]
-[timestamp] [1,]
-[timestamp] [2,]
-[timestamp] [3,]
-[timestamp] [4,]
-[timestamp] [5,]
-[timestamp] [6,]
-[timestamp] [7,]
-[timestamp] [8,]
-[timestamp] [9,]
+[timestamp] [0,]
 ----------------------------------------------------------------------
 ### Batches Started At:
 [timestamp] [0,]
-[timestamp] [1,]
-[timestamp] [2,]
-[timestamp] [3,]
-[timestamp] [4,]
-[timestamp] [5,]
-[timestamp] [6,]
-[timestamp] [7,]
-[timestamp] [8,]
-[timestamp] [9,]
 ----------------------------------------------------------------------
 ### Batches Finished At:
 [timestamp] [0,]
-[timestamp] [1,]
-[timestamp] [2,]
-[timestamp] [3,]
-[timestamp] [4,]
-[timestamp] [5,]
-[timestamp] [6,]
-[timestamp] [7,]
-[timestamp] [8,]
-[timestamp] [9,]
 ----------------------------------------------------------------------
 ### File Store:
 

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

@@ -298,11 +298,8 @@ fn test_mixed_document_addition() {
     }
     snapshot!(snapshot_index_scheduler(&index_scheduler), name: "after_registering_the_10_tasks");
 
-    // Only half of the task should've been processed since we can't autobatch replace and update together.
-    handle.advance_n_successful_batches(5);
-    snapshot!(snapshot_index_scheduler(&index_scheduler), name: "five_tasks_processed");
-
-    handle.advance_n_successful_batches(5);
+    // All tasks should've been batched and processed together since any indexing task (updates with replacements) can be batched together
+    handle.advance_n_successful_batches(1);
     snapshot!(snapshot_index_scheduler(&index_scheduler), name: "all_tasks_processed");
 
     // has everything being pushed successfully in milli?

crates/meilisearch/Cargo.toml

@@ -145,6 +145,7 @@ zip = { version = "2.2.2", optional = true }
 [features]
 default = ["meilisearch-types/all-tokenizations", "mini-dashboard"]
 swagger = ["utoipa-scalar"]
+test-ollama = []
 mini-dashboard = [
     "static-files",
     "anyhow",

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

@@ -27,6 +27,12 @@ use crate::Opt;
 /// It also generates a `configure` function that configures the routes for the settings.
 macro_rules! make_setting_routes {
     ($({route: $route:literal, update_verb: $update_verb:ident, value_type: $type:ty, err_type: $err_ty:ty, attr: $attr:ident, camelcase_attr: $camelcase_attr:literal, analytics: $analytics:ident},)*) => {
+        const _: fn(&meilisearch_types::settings::Settings<meilisearch_types::settings::Unchecked>) = |s| {
+            // This pattern match will fail at compile time if any field in Settings is not listed in the macro
+            match *s {
+                meilisearch_types::settings::Settings { $($attr: _,)* _kind: _ } => {}
+            }
+        };
         $(
             make_setting_route!($route, $update_verb, $type, $err_ty, $attr, $camelcase_attr, $analytics);
         )*
@@ -60,7 +66,7 @@ macro_rules! make_setting_routes {
 
 #[macro_export]
 macro_rules! make_setting_route {
-    ($route:literal, $update_verb:ident, $type:ty, $err_ty:ty, $attr:ident, $camelcase_attr:literal, $analytics:ident) => {
+    ($route:literal, $update_verb:ident, $type:ty, $err_type:ty, $attr:ident, $camelcase_attr:literal, $analytics:ident) => {
         pub mod $attr {
             use actix_web::web::Data;
             use actix_web::{web, HttpRequest, HttpResponse, Resource};
@@ -180,7 +186,7 @@ macro_rules! make_setting_route {
                     Data<IndexScheduler>,
                 >,
                 index_uid: actix_web::web::Path<String>,
-                body: deserr::actix_web::AwebJson<Option<$type>, $err_ty>,
+                body: deserr::actix_web::AwebJson<Option<$type>, $err_type>,
                 req: HttpRequest,
                 opt: web::Data<Opt>,
                 analytics: web::Data<Analytics>,

crates/meilisearch/tests/common/server.rs

@@ -222,6 +222,26 @@ impl Server<Shared> {
         (value, code)
     }
 
+    pub async fn list_indexes(
+        &self,
+        offset: Option<usize>,
+        limit: Option<usize>,
+    ) -> (Value, StatusCode) {
+        let (offset, limit) = (
+            offset.map(|offset| format!("offset={offset}")),
+            limit.map(|limit| format!("limit={limit}")),
+        );
+        let query_parameter = offset
+            .as_ref()
+            .zip(limit.as_ref())
+            .map(|(offset, limit)| format!("{offset}&{limit}"))
+            .or_else(|| offset.xor(limit));
+        if let Some(query_parameter) = query_parameter {
+            self.service.get(format!("/indexes?{query_parameter}")).await
+        } else {
+            self.service.get("/indexes").await
+        }
+    }
     pub async fn update_raw_index_fail(
         &self,
         uid: impl AsRef<str>,

crates/meilisearch/tests/index/get_index.rs

@@ -1,22 +1,22 @@
+use crate::json;
 use meili_snap::{json_string, snapshot};
-use serde_json::{json, Value};
+use serde_json::Value;
 
-use crate::common::Server;
+use crate::common::{shared_does_not_exists_index, Server};
 
 #[actix_rt::test]
 async fn create_and_get_index() {
-    let server = Server::new().await;
-    let index = server.index("test");
-    let (task, code) = index.create(None).await;
+    let server = Server::new_shared();
+    let index = server.unique_index();
+    let (response, code) = index.create(None).await;
 
     assert_eq!(code, 202);
 
-    index.wait_task(task.uid()).await.succeeded();
+    index.wait_task(response.uid()).await.succeeded();
 
     let (response, code) = index.get().await;
 
     assert_eq!(code, 200);
-    assert_eq!(response["uid"], "test");
     assert!(response.get("createdAt").is_some());
     assert!(response.get("updatedAt").is_some());
     assert_eq!(response["createdAt"], response["updatedAt"]);
@@ -26,13 +26,12 @@ async fn create_and_get_index() {
 
 #[actix_rt::test]
 async fn error_get_unexisting_index() {
-    let server = Server::new().await;
-    let index = server.index("test");
+    let index = shared_does_not_exists_index().await;
 
     let (response, code) = index.get().await;
 
     let expected_response = json!({
-        "message": "Index `test` not found.",
+        "message": "Index `DOES_NOT_EXISTS` not found.",
         "code": "index_not_found",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#index_not_found"
@@ -179,14 +178,14 @@ async fn get_and_paginate_indexes() {
 
 #[actix_rt::test]
 async fn get_invalid_index_uid() {
-    let server = Server::new().await;
-    let index = server.index("this is not a valid index name");
-    let (response, code) = index.get().await;
+    let server = Server::new_shared();
+    let (response, code) =
+        server.create_index_fail(json!({ "uid": "this is not a valid index name" })).await;
 
     snapshot!(code, @"400 Bad Request");
     snapshot!(json_string!(response), @r###"
     {
-      "message": "`this is not a valid index name` is not a valid index uid. Index uid can be an integer or a string containing only alphanumeric characters, hyphens (-) and underscores (_), and can not be more than 512 bytes.",
+      "message": "Invalid value at `.uid`: `this is not a valid index name` is not a valid index uid. Index uid can be an integer or a string containing only alphanumeric characters, hyphens (-) and underscores (_), and can not be more than 512 bytes.",
       "code": "invalid_index_uid",
       "type": "invalid_request",
       "link": "https://docs.meilisearch.com/errors#invalid_index_uid"

crates/meilisearch/tests/vector/mod.rs

@@ -1,4 +1,6 @@
 mod binary_quantized;
+#[cfg(feature = "test-ollama")]
+mod ollama;
 mod openai;
 mod rest;
 mod settings;

crates/meilisearch/tests/vector/ollama.rs

@@ -0,0 +1,733 @@
+//! Tests ollama embedders with the server at the location described by `MEILI_TEST_OLLAMA_SERVER` environment variable.
+
+use std::env::VarError;
+
+use meili_snap::{json_string, snapshot};
+
+use crate::common::{GetAllDocumentsOptions, Value};
+use crate::json;
+use crate::vector::get_server_vector;
+
+pub enum Endpoint {
+    /// Deprecated, undocumented endpoint
+    Embeddings,
+    /// Current endpoint
+    Embed,
+}
+
+impl Endpoint {
+    fn suffix(&self) -> &'static str {
+        match self {
+            Endpoint::Embeddings => "/api/embeddings",
+            Endpoint::Embed => "/api/embed",
+        }
+    }
+}
+
+pub enum Model {
+    Nomic,
+    AllMinilm,
+}
+
+impl Model {
+    fn name(&self) -> &'static str {
+        match self {
+            Model::Nomic => "nomic-embed-text",
+            Model::AllMinilm => "all-minilm",
+        }
+    }
+}
+
+const DOGGO_TEMPLATE: &str = r#"{%- if doc.gender == "F" -%}Une chienne nommée {{doc.name}}, née en {{doc.birthyear}}
+        {%- else -%}
+        Un chien nommé {{doc.name}}, né en {{doc.birthyear}}
+        {%- endif %}, de race {{doc.breed}}."#;
+
+fn create_ollama_config_with_template(
+    document_template: &str,
+    model: Model,
+    endpoint: Endpoint,
+) -> Option<Value> {
+    let ollama_base_url = match std::env::var("MEILI_TEST_OLLAMA_SERVER") {
+        Ok(ollama_base_url) => ollama_base_url,
+        Err(VarError::NotPresent) => return None,
+        Err(VarError::NotUnicode(s)) => panic!(
+            "`MEILI_TEST_OLLAMA_SERVER` was not properly utf-8, `{:?}`",
+            s.as_encoded_bytes()
+        ),
+    };
+
+    Some(json!({
+        "source": "ollama",
+        "url": format!("{ollama_base_url}{}", endpoint.suffix()),
+        "documentTemplate": document_template,
+        "documentTemplateMaxBytes": 8000000,
+        "model": model.name()
+    }))
+}
+
+#[actix_rt::test]
+async fn test_both_apis() {
+    let Some(embed_settings) =
+        create_ollama_config_with_template(DOGGO_TEMPLATE, Model::AllMinilm, Endpoint::Embed)
+    else {
+        panic!("Missing `MEILI_TEST_OLLAMA_SERVER` environment variable, skipping `test_both_apis` test.");
+    };
+
+    let Some(embeddings_settings) =
+        create_ollama_config_with_template(DOGGO_TEMPLATE, Model::AllMinilm, Endpoint::Embeddings)
+    else {
+        return;
+    };
+
+    let Some(nomic_embed_settings) =
+        create_ollama_config_with_template(DOGGO_TEMPLATE, Model::Nomic, Endpoint::Embed)
+    else {
+        return;
+    };
+
+    let Some(nomic_embeddings_settings) =
+        create_ollama_config_with_template(DOGGO_TEMPLATE, Model::Nomic, Endpoint::Embeddings)
+    else {
+        return;
+    };
+
+    let server = get_server_vector().await;
+
+    let index = server.index("doggo");
+
+    let (response, code) = index
+        .update_settings(json!({
+          "embedders": {
+              "embed": embed_settings,
+              "embeddings": embeddings_settings,
+              "nomic_embed": nomic_embed_settings,
+              "nomic_embeddings": nomic_embeddings_settings,
+          },
+        }))
+        .await;
+    snapshot!(code, @"202 Accepted");
+    let task = server.wait_task(response.uid()).await;
+    snapshot!(task["status"], @r###""succeeded""###);
+    let documents = json!([
+      {"id": 0, "name": "kefir", "gender": "M", "birthyear": 2023, "breed": "Patou"},
+      {"id": 1, "name": "Intel", "gender": "M", "birthyear": 2011, "breed": "Beagle"},
+      {"id": 2, "name": "Vénus", "gender": "F", "birthyear": 2003, "breed": "Jack Russel Terrier"},
+      {"id": 3, "name": "Max", "gender": "M", "birthyear": 1995, "breed": "Labrador Retriever"},
+    ]);
+    let (value, code) = index.add_documents(documents, None).await;
+    snapshot!(code, @"202 Accepted");
+    let task = index.wait_task(value.uid()).await;
+    snapshot!(task, @r###"
+    {
+      "uid": "[uid]",
+      "batchUid": "[batch_uid]",
+      "indexUid": "doggo",
+      "status": "succeeded",
+      "type": "documentAdditionOrUpdate",
+      "canceledBy": null,
+      "details": {
+        "receivedDocuments": 4,
+        "indexedDocuments": 4
+      },
+      "error": null,
+      "duration": "[duration]",
+      "enqueuedAt": "[date]",
+      "startedAt": "[date]",
+      "finishedAt": "[date]"
+    }
+    "###);
+
+    let (documents, _code) = index
+        .get_all_documents(GetAllDocumentsOptions { retrieve_vectors: true, ..Default::default() })
+        .await;
+    snapshot!(json_string!(documents, {".results.*._vectors.*.embeddings" => "[vector]"}), @r###"
+    {
+      "results": [
+        {
+          "id": 0,
+          "name": "kefir",
+          "gender": "M",
+          "birthyear": 2023,
+          "breed": "Patou",
+          "_vectors": {
+            "embed": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            },
+            "embeddings": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            },
+            "nomic_embed": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            },
+            "nomic_embeddings": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            }
+          }
+        },
+        {
+          "id": 1,
+          "name": "Intel",
+          "gender": "M",
+          "birthyear": 2011,
+          "breed": "Beagle",
+          "_vectors": {
+            "embed": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            },
+            "embeddings": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            },
+            "nomic_embed": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            },
+            "nomic_embeddings": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            }
+          }
+        },
+        {
+          "id": 2,
+          "name": "Vénus",
+          "gender": "F",
+          "birthyear": 2003,
+          "breed": "Jack Russel Terrier",
+          "_vectors": {
+            "embed": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            },
+            "embeddings": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            },
+            "nomic_embed": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            },
+            "nomic_embeddings": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            }
+          }
+        },
+        {
+          "id": 3,
+          "name": "Max",
+          "gender": "M",
+          "birthyear": 1995,
+          "breed": "Labrador Retriever",
+          "_vectors": {
+            "embed": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            },
+            "embeddings": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            },
+            "nomic_embed": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            },
+            "nomic_embeddings": {
+              "embeddings": "[vector]",
+              "regenerate": true
+            }
+          }
+        }
+      ],
+      "offset": 0,
+      "limit": 20,
+      "total": 4
+    }
+    "###);
+
+    let (response, code) = index
+        .search_post(json!({
+            "q": "chien de chasse",
+            "hybrid": {"semanticRatio": 1.0, "embedder": "embed"},
+        }))
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(json_string!(response["hits"]), @r###"
+    [
+      {
+        "id": 0,
+        "name": "kefir",
+        "gender": "M",
+        "birthyear": 2023,
+        "breed": "Patou"
+      },
+      {
+        "id": 1,
+        "name": "Intel",
+        "gender": "M",
+        "birthyear": 2011,
+        "breed": "Beagle"
+      },
+      {
+        "id": 3,
+        "name": "Max",
+        "gender": "M",
+        "birthyear": 1995,
+        "breed": "Labrador Retriever"
+      },
+      {
+        "id": 2,
+        "name": "Vénus",
+        "gender": "F",
+        "birthyear": 2003,
+        "breed": "Jack Russel Terrier"
+      }
+    ]
+    "###);
+
+    let (response, code) = index
+        .search_post(json!({
+            "q": "chien de chasse",
+            "hybrid": {"semanticRatio": 1.0, "embedder": "embeddings"},
+        }))
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(json_string!(response["hits"]), @r###"
+    [
+      {
+        "id": 0,
+        "name": "kefir",
+        "gender": "M",
+        "birthyear": 2023,
+        "breed": "Patou"
+      },
+      {
+        "id": 1,
+        "name": "Intel",
+        "gender": "M",
+        "birthyear": 2011,
+        "breed": "Beagle"
+      },
+      {
+        "id": 3,
+        "name": "Max",
+        "gender": "M",
+        "birthyear": 1995,
+        "breed": "Labrador Retriever"
+      },
+      {
+        "id": 2,
+        "name": "Vénus",
+        "gender": "F",
+        "birthyear": 2003,
+        "breed": "Jack Russel Terrier"
+      }
+    ]
+    "###);
+
+    let (response, code) = index
+        .search_post(json!({
+            "q": "petit chien",
+            "hybrid": {"semanticRatio": 1.0, "embedder": "embed"}
+        }))
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(json_string!(response["hits"]), @r###"
+    [
+      {
+        "id": 3,
+        "name": "Max",
+        "gender": "M",
+        "birthyear": 1995,
+        "breed": "Labrador Retriever"
+      },
+      {
+        "id": 1,
+        "name": "Intel",
+        "gender": "M",
+        "birthyear": 2011,
+        "breed": "Beagle"
+      },
+      {
+        "id": 0,
+        "name": "kefir",
+        "gender": "M",
+        "birthyear": 2023,
+        "breed": "Patou"
+      },
+      {
+        "id": 2,
+        "name": "Vénus",
+        "gender": "F",
+        "birthyear": 2003,
+        "breed": "Jack Russel Terrier"
+      }
+    ]
+    "###);
+
+    let (response, code) = index
+        .search_post(json!({
+            "q": "petit chien",
+            "hybrid": {"semanticRatio": 1.0, "embedder": "embeddings"}
+        }))
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(json_string!(response["hits"]), @r###"
+    [
+      {
+        "id": 3,
+        "name": "Max",
+        "gender": "M",
+        "birthyear": 1995,
+        "breed": "Labrador Retriever"
+      },
+      {
+        "id": 1,
+        "name": "Intel",
+        "gender": "M",
+        "birthyear": 2011,
+        "breed": "Beagle"
+      },
+      {
+        "id": 0,
+        "name": "kefir",
+        "gender": "M",
+        "birthyear": 2023,
+        "breed": "Patou"
+      },
+      {
+        "id": 2,
+        "name": "Vénus",
+        "gender": "F",
+        "birthyear": 2003,
+        "breed": "Jack Russel Terrier"
+      }
+    ]
+    "###);
+
+    let (response, code) = index
+        .search_post(json!({
+            "q": "grand chien de berger des montagnes",
+            "hybrid": {"semanticRatio": 1.0, "embedder": "embed"}
+        }))
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(json_string!(response["hits"]), @r###"
+    [
+      {
+        "id": 0,
+        "name": "kefir",
+        "gender": "M",
+        "birthyear": 2023,
+        "breed": "Patou"
+      },
+      {
+        "id": 1,
+        "name": "Intel",
+        "gender": "M",
+        "birthyear": 2011,
+        "breed": "Beagle"
+      },
+      {
+        "id": 3,
+        "name": "Max",
+        "gender": "M",
+        "birthyear": 1995,
+        "breed": "Labrador Retriever"
+      },
+      {
+        "id": 2,
+        "name": "Vénus",
+        "gender": "F",
+        "birthyear": 2003,
+        "breed": "Jack Russel Terrier"
+      }
+    ]
+    "###);
+
+    let (response, code) = index
+        .search_post(json!({
+            "q": "grand chien de berger des montagnes",
+            "hybrid": {"semanticRatio": 1.0, "embedder": "embeddings"}
+        }))
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(json_string!(response["hits"]), @r###"
+    [
+      {
+        "id": 0,
+        "name": "kefir",
+        "gender": "M",
+        "birthyear": 2023,
+        "breed": "Patou"
+      },
+      {
+        "id": 1,
+        "name": "Intel",
+        "gender": "M",
+        "birthyear": 2011,
+        "breed": "Beagle"
+      },
+      {
+        "id": 3,
+        "name": "Max",
+        "gender": "M",
+        "birthyear": 1995,
+        "breed": "Labrador Retriever"
+      },
+      {
+        "id": 2,
+        "name": "Vénus",
+        "gender": "F",
+        "birthyear": 2003,
+        "breed": "Jack Russel Terrier"
+      }
+    ]
+    "###);
+
+    let (response, code) = index
+        .search_post(json!({
+            "q": "chien de chasse",
+            "hybrid": {"semanticRatio": 1.0, "embedder": "nomic_embed"},
+        }))
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(json_string!(response["hits"]), @r###"
+    [
+      {
+        "id": 0,
+        "name": "kefir",
+        "gender": "M",
+        "birthyear": 2023,
+        "breed": "Patou"
+      },
+      {
+        "id": 2,
+        "name": "Vénus",
+        "gender": "F",
+        "birthyear": 2003,
+        "breed": "Jack Russel Terrier"
+      },
+      {
+        "id": 1,
+        "name": "Intel",
+        "gender": "M",
+        "birthyear": 2011,
+        "breed": "Beagle"
+      },
+      {
+        "id": 3,
+        "name": "Max",
+        "gender": "M",
+        "birthyear": 1995,
+        "breed": "Labrador Retriever"
+      }
+    ]
+    "###);
+
+    let (response, code) = index
+        .search_post(json!({
+            "q": "chien de chasse",
+            "hybrid": {"semanticRatio": 1.0, "embedder": "nomic_embeddings"},
+        }))
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(json_string!(response["hits"]), @r###"
+    [
+      {
+        "id": 0,
+        "name": "kefir",
+        "gender": "M",
+        "birthyear": 2023,
+        "breed": "Patou"
+      },
+      {
+        "id": 2,
+        "name": "Vénus",
+        "gender": "F",
+        "birthyear": 2003,
+        "breed": "Jack Russel Terrier"
+      },
+      {
+        "id": 1,
+        "name": "Intel",
+        "gender": "M",
+        "birthyear": 2011,
+        "breed": "Beagle"
+      },
+      {
+        "id": 3,
+        "name": "Max",
+        "gender": "M",
+        "birthyear": 1995,
+        "breed": "Labrador Retriever"
+      }
+    ]
+    "###);
+
+    let (response, code) = index
+        .search_post(json!({
+            "q": "petit chien",
+            "hybrid": {"semanticRatio": 1.0, "embedder": "nomic_embed"}
+        }))
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(json_string!(response["hits"]), @r###"
+    [
+      {
+        "id": 0,
+        "name": "kefir",
+        "gender": "M",
+        "birthyear": 2023,
+        "breed": "Patou"
+      },
+      {
+        "id": 3,
+        "name": "Max",
+        "gender": "M",
+        "birthyear": 1995,
+        "breed": "Labrador Retriever"
+      },
+      {
+        "id": 2,
+        "name": "Vénus",
+        "gender": "F",
+        "birthyear": 2003,
+        "breed": "Jack Russel Terrier"
+      },
+      {
+        "id": 1,
+        "name": "Intel",
+        "gender": "M",
+        "birthyear": 2011,
+        "breed": "Beagle"
+      }
+    ]
+    "###);
+
+    let (response, code) = index
+        .search_post(json!({
+            "q": "petit chien",
+            "hybrid": {"semanticRatio": 1.0, "embedder": "nomic_embeddings"}
+        }))
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(json_string!(response["hits"]), @r###"
+    [
+      {
+        "id": 0,
+        "name": "kefir",
+        "gender": "M",
+        "birthyear": 2023,
+        "breed": "Patou"
+      },
+      {
+        "id": 3,
+        "name": "Max",
+        "gender": "M",
+        "birthyear": 1995,
+        "breed": "Labrador Retriever"
+      },
+      {
+        "id": 2,
+        "name": "Vénus",
+        "gender": "F",
+        "birthyear": 2003,
+        "breed": "Jack Russel Terrier"
+      },
+      {
+        "id": 1,
+        "name": "Intel",
+        "gender": "M",
+        "birthyear": 2011,
+        "breed": "Beagle"
+      }
+    ]
+    "###);
+
+    let (response, code) = index
+        .search_post(json!({
+            "q": "grand chien de berger des montagnes",
+            "hybrid": {"semanticRatio": 1.0, "embedder": "nomic_embed"}
+        }))
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(json_string!(response["hits"]), @r###"
+    [
+      {
+        "id": 0,
+        "name": "kefir",
+        "gender": "M",
+        "birthyear": 2023,
+        "breed": "Patou"
+      },
+      {
+        "id": 3,
+        "name": "Max",
+        "gender": "M",
+        "birthyear": 1995,
+        "breed": "Labrador Retriever"
+      },
+      {
+        "id": 2,
+        "name": "Vénus",
+        "gender": "F",
+        "birthyear": 2003,
+        "breed": "Jack Russel Terrier"
+      },
+      {
+        "id": 1,
+        "name": "Intel",
+        "gender": "M",
+        "birthyear": 2011,
+        "breed": "Beagle"
+      }
+    ]
+    "###);
+
+    let (response, code) = index
+        .search_post(json!({
+            "q": "grand chien de berger des montagnes",
+            "hybrid": {"semanticRatio": 1.0, "embedder": "nomic_embeddings"}
+        }))
+        .await;
+    snapshot!(code, @"200 OK");
+    snapshot!(json_string!(response["hits"]), @r###"
+    [
+      {
+        "id": 0,
+        "name": "kefir",
+        "gender": "M",
+        "birthyear": 2023,
+        "breed": "Patou"
+      },
+      {
+        "id": 3,
+        "name": "Max",
+        "gender": "M",
+        "birthyear": 1995,
+        "breed": "Labrador Retriever"
+      },
+      {
+        "id": 2,
+        "name": "Vénus",
+        "gender": "F",
+        "birthyear": 2003,
+        "breed": "Jack Russel Terrier"
+      },
+      {
+        "id": 1,
+        "name": "Intel",
+        "gender": "M",
+        "birthyear": 2011,
+        "breed": "Beagle"
+      }
+    ]
+    "###);
+}

crates/milli/src/index.rs

@@ -1839,9 +1839,15 @@ pub(crate) mod tests {
 
             let embedders =
                 InnerIndexSettings::from_index(&self.inner, &rtxn, None)?.embedding_configs;
-            let mut indexer =
-                indexer::DocumentOperation::new(self.index_documents_config.update_method);
-            indexer.add_documents(&documents).unwrap();
+            let mut indexer = indexer::DocumentOperation::new();
+            match self.index_documents_config.update_method {
+                IndexDocumentsMethod::ReplaceDocuments => {
+                    indexer.replace_documents(&documents).unwrap()
+                }
+                IndexDocumentsMethod::UpdateDocuments => {
+                    indexer.update_documents(&documents).unwrap()
+                }
+            }
 
             let indexer_alloc = Bump::new();
             let (document_changes, operation_stats, primary_key) = indexer.into_changes(
@@ -1928,8 +1934,7 @@ pub(crate) mod tests {
             let embedders =
                 InnerIndexSettings::from_index(&self.inner, &rtxn, None)?.embedding_configs;
 
-            let mut indexer =
-                indexer::DocumentOperation::new(self.index_documents_config.update_method);
+            let mut indexer = indexer::DocumentOperation::new();
             let external_document_ids: Vec<_> =
                 external_document_ids.iter().map(AsRef::as_ref).collect();
             indexer.delete_documents(external_document_ids.as_slice());
@@ -2006,13 +2011,13 @@ pub(crate) mod tests {
         let mut new_fields_ids_map = db_fields_ids_map.clone();
 
         let embedders = EmbeddingConfigs::default();
-        let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+        let mut indexer = indexer::DocumentOperation::new();
         let payload = documents!([
             { "id": 1, "name": "kevin" },
             { "id": 2, "name": "bob", "age": 20 },
             { "id": 2, "name": "bob", "age": 20 },
         ]);
-        indexer.add_documents(&payload).unwrap();
+        indexer.replace_documents(&payload).unwrap();
 
         let indexer_alloc = Bump::new();
         let (document_changes, _operation_stats, primary_key) = indexer

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

@@ -7,7 +7,7 @@ use maplit::{btreemap, hashset};
 
 use crate::progress::Progress;
 use crate::update::new::indexer;
-use crate::update::{IndexDocumentsMethod, IndexerConfig, Settings};
+use crate::update::{IndexerConfig, Settings};
 use crate::vector::EmbeddingConfigs;
 use crate::{db_snap, Criterion, Index};
 pub const CONTENT: &str = include_str!("../../../../tests/assets/test_set.ndjson");
@@ -55,7 +55,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
     let mut new_fields_ids_map = db_fields_ids_map.clone();
 
     let embedders = EmbeddingConfigs::default();
-    let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+    let mut indexer = indexer::DocumentOperation::new();
 
     let mut file = tempfile::tempfile().unwrap();
     file.write_all(CONTENT.as_bytes()).unwrap();
@@ -63,7 +63,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
     let payload = unsafe { memmap2::Mmap::map(&file).unwrap() };
 
     // index documents
-    indexer.add_documents(&payload).unwrap();
+    indexer.replace_documents(&payload).unwrap();
 
     let indexer_alloc = Bump::new();
     let (document_changes, operation_stats, primary_key) = indexer

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

@@ -771,7 +771,7 @@ mod tests {
     use crate::search::TermsMatchingStrategy;
     use crate::update::new::indexer;
     use crate::update::Setting;
-    use crate::{db_snap, Filter, Search, UserError};
+    use crate::{all_obkv_to_json, db_snap, Filter, Search, UserError};
 
     #[test]
     fn simple_document_replacement() {
@@ -1951,11 +1951,11 @@ mod tests {
         let db_fields_ids_map = index.inner.fields_ids_map(&rtxn).unwrap();
         let mut new_fields_ids_map = db_fields_ids_map.clone();
 
-        let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
-        indexer.add_documents(&doc1).unwrap();
-        indexer.add_documents(&doc2).unwrap();
-        indexer.add_documents(&doc3).unwrap();
-        indexer.add_documents(&doc4).unwrap();
+        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();
 
         let indexer_alloc = Bump::new();
         let (_document_changes, operation_stats, _primary_key) = indexer
@@ -1974,6 +1974,174 @@ mod tests {
         assert_eq!(operation_stats.iter().filter(|ps| ps.error.is_some()).count(), 3);
     }
 
+    #[test]
+    fn mixing_documents_replace_with_updates() {
+        let index = TempIndex::new_with_map_size(4096 * 100);
+
+        let doc1 = documents! {[{
+            "id": 1,
+            "title": "asdsad",
+            "description": "Wat wat wat, wat"
+        }]};
+
+        let doc2 = documents! {[{
+            "id": 1,
+            "title": "something",
+        }]};
+
+        let doc3 = documents! {[{
+            "id": 1,
+            "title": "another something",
+        }]};
+
+        let doc4 = documents! {[{
+            "id": 1,
+            "description": "This is it!",
+        }]};
+
+        let rtxn = index.inner.read_txn().unwrap();
+        let db_fields_ids_map = index.inner.fields_ids_map(&rtxn).unwrap();
+        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();
+
+        let indexer_alloc = Bump::new();
+        let (document_changes, operation_stats, primary_key) = indexer
+            .into_changes(
+                &indexer_alloc,
+                &index.inner,
+                &rtxn,
+                None,
+                &mut new_fields_ids_map,
+                &|| false,
+                Progress::default(),
+            )
+            .unwrap();
+
+        assert_eq!(operation_stats.iter().filter(|ps| ps.error.is_none()).count(), 4);
+
+        let mut wtxn = index.write_txn().unwrap();
+        indexer::index(
+            &mut wtxn,
+            &index.inner,
+            &crate::ThreadPoolNoAbortBuilder::new().build().unwrap(),
+            index.indexer_config.grenad_parameters(),
+            &db_fields_ids_map,
+            new_fields_ids_map,
+            primary_key,
+            &document_changes,
+            EmbeddingConfigs::default(),
+            &|| false,
+            &Progress::default(),
+        )
+        .unwrap();
+        wtxn.commit().unwrap();
+
+        let rtxn = index.read_txn().unwrap();
+        let obkv = index.document(&rtxn, 0).unwrap();
+        let fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
+
+        let json_document = all_obkv_to_json(obkv, &fields_ids_map).unwrap();
+        let expected = serde_json::json!({
+            "id": 1,
+            "title": "another something",
+            "description": "This is it!",
+        });
+        let expected = expected.as_object().unwrap();
+        assert_eq!(&json_document, expected);
+    }
+
+    #[test]
+    fn mixing_documents_replace_with_updates_even_more() {
+        let index = TempIndex::new_with_map_size(4096 * 100);
+
+        let doc1 = documents! {[{
+            "id": 1,
+            "title": "asdsad",
+            "description": "Wat wat wat, wat"
+        }]};
+
+        let doc2 = documents! {[{
+            "id": 1,
+            "title": "something",
+        }]};
+
+        let doc3 = documents! {[{
+            "id": 1,
+            "title": "another something",
+        }]};
+
+        let doc4 = documents! {[{
+            "id": 1,
+            "title": "Woooof",
+        }]};
+
+        let doc5 = documents! {[{
+            "id": 1,
+            "description": "This is it!",
+        }]};
+
+        let rtxn = index.inner.read_txn().unwrap();
+        let db_fields_ids_map = index.inner.fields_ids_map(&rtxn).unwrap();
+        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();
+
+        let indexer_alloc = Bump::new();
+        let (document_changes, operation_stats, primary_key) = indexer
+            .into_changes(
+                &indexer_alloc,
+                &index.inner,
+                &rtxn,
+                None,
+                &mut new_fields_ids_map,
+                &|| false,
+                Progress::default(),
+            )
+            .unwrap();
+
+        assert_eq!(operation_stats.iter().filter(|ps| ps.error.is_none()).count(), 5);
+
+        let mut wtxn = index.write_txn().unwrap();
+        indexer::index(
+            &mut wtxn,
+            &index.inner,
+            &crate::ThreadPoolNoAbortBuilder::new().build().unwrap(),
+            index.indexer_config.grenad_parameters(),
+            &db_fields_ids_map,
+            new_fields_ids_map,
+            primary_key,
+            &document_changes,
+            EmbeddingConfigs::default(),
+            &|| false,
+            &Progress::default(),
+        )
+        .unwrap();
+        wtxn.commit().unwrap();
+
+        let rtxn = index.read_txn().unwrap();
+        let obkv = index.document(&rtxn, 0).unwrap();
+        let fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
+
+        let json_document = all_obkv_to_json(obkv, &fields_ids_map).unwrap();
+        let expected = serde_json::json!({
+            "id": 1,
+            "title": "Woooof",
+            "description": "This is it!",
+        });
+        let expected = expected.as_object().unwrap();
+        assert_eq!(&json_document, expected);
+    }
+
     #[test]
     fn primary_key_must_not_contain_whitespace() {
         let index = TempIndex::new();
@@ -2112,8 +2280,8 @@ mod tests {
 
         let indexer_alloc = Bump::new();
         let embedders = EmbeddingConfigs::default();
-        let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
-        indexer.add_documents(&documents).unwrap();
+        let mut indexer = indexer::DocumentOperation::new();
+        indexer.replace_documents(&documents).unwrap();
         indexer.delete_documents(&["2"]);
         let (document_changes, _operation_stats, primary_key) = indexer
             .into_changes(
@@ -2165,14 +2333,14 @@ mod tests {
             { "id": 2, "doggo": { "name": "bob", "age": 20 } },
             { "id": 3, "name": "jean", "age": 25 },
         ]);
-        let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::UpdateDocuments);
-        indexer.add_documents(&documents).unwrap();
+        let mut indexer = indexer::DocumentOperation::new();
+        indexer.update_documents(&documents).unwrap();
 
         let documents = documents!([
             { "id": 2, "catto": "jorts" },
             { "id": 3, "legs": 4 },
         ]);
-        indexer.add_documents(&documents).unwrap();
+        indexer.update_documents(&documents).unwrap();
         indexer.delete_documents(&["1", "2"]);
 
         let indexer_alloc = Bump::new();
@@ -2227,8 +2395,8 @@ mod tests {
         ]);
         let indexer_alloc = Bump::new();
         let embedders = EmbeddingConfigs::default();
-        let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::UpdateDocuments);
-        indexer.add_documents(&documents).unwrap();
+        let mut indexer = indexer::DocumentOperation::new();
+        indexer.update_documents(&documents).unwrap();
 
         let (document_changes, _operation_stats, primary_key) = indexer
             .into_changes(
@@ -2278,8 +2446,8 @@ mod tests {
         ]);
         let indexer_alloc = Bump::new();
         let embedders = EmbeddingConfigs::default();
-        let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::UpdateDocuments);
-        indexer.add_documents(&documents).unwrap();
+        let mut indexer = indexer::DocumentOperation::new();
+        indexer.update_documents(&documents).unwrap();
         indexer.delete_documents(&["1", "2"]);
 
         let (document_changes, _operation_stats, primary_key) = indexer
@@ -2327,14 +2495,14 @@ mod tests {
 
         let indexer_alloc = Bump::new();
         let embedders = EmbeddingConfigs::default();
-        let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::UpdateDocuments);
+        let mut indexer = indexer::DocumentOperation::new();
         indexer.delete_documents(&["1", "2"]);
 
         let documents = documents!([
             { "id": 2, "doggo": { "name": "jean", "age": 20 } },
             { "id": 3, "name": "bob", "age": 25 },
         ]);
-        indexer.add_documents(&documents).unwrap();
+        indexer.update_documents(&documents).unwrap();
 
         let (document_changes, _operation_stats, primary_key) = indexer
             .into_changes(
@@ -2382,7 +2550,7 @@ mod tests {
 
         let indexer_alloc = Bump::new();
         let embedders = EmbeddingConfigs::default();
-        let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::UpdateDocuments);
+        let mut indexer = indexer::DocumentOperation::new();
 
         indexer.delete_documents(&["1", "2", "1", "2"]);
 
@@ -2391,7 +2559,7 @@ mod tests {
             { "id": 2, "doggo": { "name": "jean", "age": 20 } },
             { "id": 3, "name": "bob", "age": 25 },
         ]);
-        indexer.add_documents(&documents).unwrap();
+        indexer.update_documents(&documents).unwrap();
 
         indexer.delete_documents(&["1", "2", "1", "2"]);
 
@@ -2440,12 +2608,12 @@ mod tests {
 
         let indexer_alloc = Bump::new();
         let embedders = EmbeddingConfigs::default();
-        let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::UpdateDocuments);
+        let mut indexer = indexer::DocumentOperation::new();
 
         let documents = documents!([
             { "id": 1, "doggo": "kevin" },
         ]);
-        indexer.add_documents(&documents).unwrap();
+        indexer.update_documents(&documents).unwrap();
 
         let (document_changes, _operation_stats, primary_key) = indexer
             .into_changes(
@@ -2489,7 +2657,7 @@ mod tests {
 
         let indexer_alloc = Bump::new();
         let embedders = EmbeddingConfigs::default();
-        let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+        let mut indexer = indexer::DocumentOperation::new();
 
         indexer.delete_documents(&["1"]);
 
@@ -2497,7 +2665,7 @@ mod tests {
             { "id": 1, "catto": "jorts" },
         ]);
 
-        indexer.add_documents(&documents).unwrap();
+        indexer.replace_documents(&documents).unwrap();
 
         let (document_changes, _operation_stats, primary_key) = indexer
             .into_changes(
@@ -2683,14 +2851,14 @@ mod tests {
 
         let indexer_alloc = Bump::new();
         let embedders = EmbeddingConfigs::default();
-        let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+        let mut indexer = indexer::DocumentOperation::new();
 
         // OP
 
         let documents = documents!([
             { "id": 1, "doggo": "bernese" },
         ]);
-        indexer.add_documents(&documents).unwrap();
+        indexer.replace_documents(&documents).unwrap();
 
         // FINISHING
         let (document_changes, _operation_stats, primary_key) = indexer
@@ -2743,14 +2911,14 @@ mod tests {
 
         let indexer_alloc = Bump::new();
         let embedders = EmbeddingConfigs::default();
-        let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+        let mut indexer = indexer::DocumentOperation::new();
 
         indexer.delete_documents(&["1"]);
 
         let documents = documents!([
             { "id": 0, "catto": "jorts" },
         ]);
-        indexer.add_documents(&documents).unwrap();
+        indexer.replace_documents(&documents).unwrap();
 
         let (document_changes, _operation_stats, primary_key) = indexer
             .into_changes(
@@ -2801,12 +2969,12 @@ mod tests {
 
         let indexer_alloc = Bump::new();
         let embedders = EmbeddingConfigs::default();
-        let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+        let mut indexer = indexer::DocumentOperation::new();
 
         let documents = documents!([
             { "id": 1, "catto": "jorts" },
         ]);
-        indexer.add_documents(&documents).unwrap();
+        indexer.replace_documents(&documents).unwrap();
 
         let (document_changes, _operation_stats, primary_key) = indexer
             .into_changes(

crates/milli/src/update/new/document_change.rs

@@ -27,7 +27,7 @@ pub struct Update<'doc> {
     docid: DocumentId,
     external_document_id: &'doc str,
     new: Versions<'doc>,
-    has_deletion: bool,
+    from_scratch: bool,
 }
 
 pub struct Insertion<'doc> {
@@ -109,9 +109,9 @@ impl<'doc> Update<'doc> {
         docid: DocumentId,
         external_document_id: &'doc str,
         new: Versions<'doc>,
-        has_deletion: bool,
+        from_scratch: bool,
     ) -> Self {
-        Update { docid, new, external_document_id, has_deletion }
+        Update { docid, new, external_document_id, from_scratch }
     }
 
     pub fn docid(&self) -> DocumentId {
@@ -154,7 +154,7 @@ impl<'doc> Update<'doc> {
         index: &'t Index,
         mapper: &'t Mapper,
     ) -> Result<MergedDocument<'_, 'doc, 't, Mapper>> {
-        if self.has_deletion {
+        if self.from_scratch {
             Ok(MergedDocument::without_db(DocumentFromVersions::new(&self.new)))
         } else {
             MergedDocument::with_db(
@@ -207,8 +207,8 @@ impl<'doc> Update<'doc> {
             cached_current = Some(current);
         }
 
-        if !self.has_deletion {
-            // no field deletion, so fields that don't appear in `updated` cannot have changed
+        if !self.from_scratch {
+            // no field deletion or update, so fields that don't appear in `updated` cannot have changed
             return Ok(changed);
         }
 
@@ -257,7 +257,7 @@ impl<'doc> Update<'doc> {
         doc_alloc: &'doc Bump,
         embedders: &'doc EmbeddingConfigs,
     ) -> Result<Option<MergedVectorDocument<'doc>>> {
-        if self.has_deletion {
+        if self.from_scratch {
             MergedVectorDocument::without_db(
                 self.external_document_id,
                 &self.new,

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

@@ -23,25 +23,39 @@ use crate::update::new::{Deletion, Insertion, Update};
 use crate::update::{AvailableIds, IndexDocumentsMethod};
 use crate::{DocumentId, Error, FieldsIdsMap, Index, InternalError, Result, UserError};
 
+#[derive(Default)]
 pub struct DocumentOperation<'pl> {
     operations: Vec<Payload<'pl>>,
-    method: MergeMethod,
 }
 
 impl<'pl> DocumentOperation<'pl> {
-    pub fn new(method: IndexDocumentsMethod) -> Self {
-        Self { operations: Default::default(), method: MergeMethod::from(method) }
+    pub fn new() -> Self {
+        Self { operations: Default::default() }
     }
 
-    /// TODO please give me a type
+    /// Append a replacement of documents.
+    ///
     /// The payload is expected to be in the NDJSON format
-    pub fn add_documents(&mut self, payload: &'pl Mmap) -> Result<()> {
+    pub fn replace_documents(&mut self, payload: &'pl Mmap) -> Result<()> {
         #[cfg(unix)]
         payload.advise(memmap2::Advice::Sequential)?;
-        self.operations.push(Payload::Addition(&payload[..]));
+        self.operations.push(Payload::Replace(&payload[..]));
         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<()> {
+        #[cfg(unix)]
+        payload.advise(memmap2::Advice::Sequential)?;
+        self.operations.push(Payload::Update(&payload[..]));
+        Ok(())
+    }
+
+    /// Append a deletion of documents IDs.
+    ///
+    /// The list is a set of external documents IDs.
     pub fn delete_documents(&mut self, to_delete: &'pl [&'pl str]) {
         self.operations.push(Payload::Deletion(to_delete))
     }
@@ -62,7 +76,7 @@ impl<'pl> DocumentOperation<'pl> {
         MSP: Fn() -> bool,
     {
         progress.update_progress(IndexingStep::PreparingPayloads);
-        let Self { operations, method } = self;
+        let Self { operations } = self;
 
         let documents_ids = index.documents_ids(rtxn)?;
         let mut operations_stats = Vec::new();
@@ -82,7 +96,7 @@ impl<'pl> DocumentOperation<'pl> {
 
             let mut bytes = 0;
             let result = match operation {
-                Payload::Addition(payload) => extract_addition_payload_changes(
+                Payload::Replace(payload) => extract_addition_payload_changes(
                     indexer,
                     index,
                     rtxn,
@@ -92,7 +106,20 @@ impl<'pl> DocumentOperation<'pl> {
                     &mut available_docids,
                     &mut bytes,
                     &docids_version_offsets,
-                    method,
+                    IndexDocumentsMethod::ReplaceDocuments,
+                    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,
                     payload,
                 ),
                 Payload::Deletion(to_delete) => extract_deletion_payload_changes(
@@ -100,7 +127,6 @@ impl<'pl> DocumentOperation<'pl> {
                     rtxn,
                     &mut available_docids,
                     &docids_version_offsets,
-                    method,
                     to_delete,
                 ),
             };
@@ -126,20 +152,15 @@ impl<'pl> DocumentOperation<'pl> {
             docids_version_offsets.drain().collect_in(indexer);
 
         // Reorder the offsets to make sure we iterate on the file sequentially
-        // And finally sort them
-        docids_version_offsets.sort_unstable_by_key(|(_, po)| method.sort_key(&po.operations));
+        // And finally sort them. This clearly speeds up reading the update files.
+        docids_version_offsets
+            .sort_unstable_by_key(|(_, po)| first_update_pointer(&po.operations).unwrap_or(0));
 
         let docids_version_offsets = docids_version_offsets.into_bump_slice();
         Ok((DocumentOperationChanges { docids_version_offsets }, operations_stats, primary_key))
     }
 }
 
-impl Default for DocumentOperation<'_> {
-    fn default() -> Self {
-        DocumentOperation::new(IndexDocumentsMethod::default())
-    }
-}
-
 #[allow(clippy::too_many_arguments)]
 fn extract_addition_payload_changes<'r, 'pl: 'r>(
     indexer: &'pl Bump,
@@ -151,9 +172,11 @@ fn extract_addition_payload_changes<'r, 'pl: 'r>(
     available_docids: &mut AvailableIds,
     bytes: &mut u64,
     main_docids_version_offsets: &hashbrown::HashMap<&'pl str, PayloadOperations<'pl>>,
-    method: MergeMethod,
+    method: IndexDocumentsMethod,
     payload: &'pl [u8],
 ) -> Result<hashbrown::HashMap<&'pl str, PayloadOperations<'pl>>> {
+    use IndexDocumentsMethod::{ReplaceDocuments, UpdateDocuments};
+
     let mut new_docids_version_offsets = hashbrown::HashMap::<&str, PayloadOperations<'pl>>::new();
 
     let mut previous_offset = 0;
@@ -204,48 +227,82 @@ fn extract_addition_payload_changes<'r, 'pl: 'r>(
             None => {
                 match index.external_documents_ids().get(rtxn, external_id) {
                     Ok(Some(docid)) => match new_docids_version_offsets.entry(external_id) {
-                        Entry::Occupied(mut entry) => {
-                            entry.get_mut().push_addition(document_offset)
-                        }
+                        Entry::Occupied(mut entry) => match method {
+                            ReplaceDocuments => entry.get_mut().push_replacement(document_offset),
+                            UpdateDocuments => entry.get_mut().push_update(document_offset),
+                        },
                         Entry::Vacant(entry) => {
-                            entry.insert(PayloadOperations::new_addition(
-                                method,
-                                docid,
-                                false, // is new
-                                document_offset,
-                            ));
+                            match method {
+                                ReplaceDocuments => {
+                                    entry.insert(PayloadOperations::new_replacement(
+                                        docid,
+                                        false, // is new
+                                        document_offset,
+                                    ));
+                                }
+                                UpdateDocuments => {
+                                    entry.insert(PayloadOperations::new_update(
+                                        docid,
+                                        false, // is new
+                                        document_offset,
+                                    ));
+                                }
+                            }
                         }
                     },
                     Ok(None) => match new_docids_version_offsets.entry(external_id) {
-                        Entry::Occupied(mut entry) => {
-                            entry.get_mut().push_addition(document_offset)
-                        }
+                        Entry::Occupied(mut entry) => match method {
+                            ReplaceDocuments => entry.get_mut().push_replacement(document_offset),
+                            UpdateDocuments => entry.get_mut().push_update(document_offset),
+                        },
                         Entry::Vacant(entry) => {
                             let docid = match available_docids.next() {
                                 Some(docid) => docid,
                                 None => return Err(UserError::DocumentLimitReached.into()),
                             };
-                            entry.insert(PayloadOperations::new_addition(
-                                method,
-                                docid,
-                                true, // is new
-                                document_offset,
-                            ));
+
+                            match method {
+                                ReplaceDocuments => {
+                                    entry.insert(PayloadOperations::new_replacement(
+                                        docid,
+                                        true, // is new
+                                        document_offset,
+                                    ));
+                                }
+                                UpdateDocuments => {
+                                    entry.insert(PayloadOperations::new_update(
+                                        docid,
+                                        true, // is new
+                                        document_offset,
+                                    ));
+                                }
+                            }
                         }
                     },
                     Err(e) => return Err(e.into()),
                 }
             }
             Some(payload_operations) => match new_docids_version_offsets.entry(external_id) {
-                Entry::Occupied(mut entry) => entry.get_mut().push_addition(document_offset),
-                Entry::Vacant(entry) => {
-                    entry.insert(PayloadOperations::new_addition(
-                        method,
-                        payload_operations.docid,
-                        payload_operations.is_new,
-                        document_offset,
-                    ));
-                }
+                Entry::Occupied(mut entry) => match method {
+                    ReplaceDocuments => entry.get_mut().push_replacement(document_offset),
+                    UpdateDocuments => entry.get_mut().push_update(document_offset),
+                },
+                Entry::Vacant(entry) => match method {
+                    ReplaceDocuments => {
+                        entry.insert(PayloadOperations::new_replacement(
+                            payload_operations.docid,
+                            payload_operations.is_new,
+                            document_offset,
+                        ));
+                    }
+                    UpdateDocuments => {
+                        entry.insert(PayloadOperations::new_update(
+                            payload_operations.docid,
+                            payload_operations.is_new,
+                            document_offset,
+                        ));
+                    }
+                },
             },
         }
 
@@ -278,7 +335,6 @@ fn extract_deletion_payload_changes<'s, 'pl: 's>(
     rtxn: &RoTxn,
     available_docids: &mut AvailableIds,
     main_docids_version_offsets: &hashbrown::HashMap<&'s str, PayloadOperations<'pl>>,
-    method: MergeMethod,
     to_delete: &'pl [&'pl str],
 ) -> Result<hashbrown::HashMap<&'s str, PayloadOperations<'pl>>> {
     let mut new_docids_version_offsets = hashbrown::HashMap::<&str, PayloadOperations<'pl>>::new();
@@ -292,7 +348,7 @@ fn extract_deletion_payload_changes<'s, 'pl: 's>(
                             Entry::Occupied(mut entry) => entry.get_mut().push_deletion(),
                             Entry::Vacant(entry) => {
                                 entry.insert(PayloadOperations::new_deletion(
-                                    method, docid, false, // is new
+                                    docid, false, // is new
                                 ));
                             }
                         }
@@ -306,7 +362,7 @@ fn extract_deletion_payload_changes<'s, 'pl: 's>(
                             Entry::Occupied(mut entry) => entry.get_mut().push_deletion(),
                             Entry::Vacant(entry) => {
                                 entry.insert(PayloadOperations::new_deletion(
-                                    method, docid, true, // is new
+                                    docid, true, // is new
                                 ));
                             }
                         }
@@ -318,7 +374,6 @@ fn extract_deletion_payload_changes<'s, 'pl: 's>(
                 Entry::Occupied(mut entry) => entry.get_mut().push_deletion(),
                 Entry::Vacant(entry) => {
                     entry.insert(PayloadOperations::new_deletion(
-                        method,
                         payload_operations.docid,
                         payload_operations.is_new,
                     ));
@@ -369,13 +424,7 @@ impl<'pl> DocumentChanges<'pl> for DocumentOperationChanges<'pl> {
         'pl: 'doc,
     {
         let (external_doc, payload_operations) = item;
-        payload_operations.merge_method.merge(
-            payload_operations.docid,
-            external_doc,
-            payload_operations.is_new,
-            &context.doc_alloc,
-            &payload_operations.operations[..],
-        )
+        payload_operations.merge(external_doc, &context.doc_alloc)
     }
 
     fn len(&self) -> usize {
@@ -388,7 +437,8 @@ pub struct DocumentOperationChanges<'pl> {
 }
 
 pub enum Payload<'pl> {
-    Addition(&'pl [u8]),
+    Replace(&'pl [u8]),
+    Update(&'pl [u8]),
     Deletion(&'pl [&'pl str]),
 }
 
@@ -405,31 +455,30 @@ pub struct PayloadOperations<'pl> {
     pub is_new: bool,
     /// The operations to perform, in order, on this document.
     pub operations: Vec<InnerDocOp<'pl>>,
-    /// The merge method we are using to merge payloads and documents.
-    merge_method: MergeMethod,
 }
 
 impl<'pl> PayloadOperations<'pl> {
-    fn new_deletion(merge_method: MergeMethod, docid: DocumentId, is_new: bool) -> Self {
-        Self { docid, is_new, operations: vec![InnerDocOp::Deletion], merge_method }
+    fn new_replacement(docid: DocumentId, is_new: bool, offset: DocumentOffset<'pl>) -> Self {
+        Self { docid, is_new, operations: vec![InnerDocOp::Replace(offset)] }
     }
 
-    fn new_addition(
-        merge_method: MergeMethod,
-        docid: DocumentId,
-        is_new: bool,
-        offset: DocumentOffset<'pl>,
-    ) -> Self {
-        Self { docid, is_new, operations: vec![InnerDocOp::Addition(offset)], merge_method }
+    fn new_update(docid: DocumentId, is_new: bool, offset: DocumentOffset<'pl>) -> Self {
+        Self { docid, is_new, operations: vec![InnerDocOp::Update(offset)] }
+    }
+
+    fn new_deletion(docid: DocumentId, is_new: bool) -> Self {
+        Self { docid, is_new, operations: vec![InnerDocOp::Deletion] }
     }
 }
 
 impl<'pl> PayloadOperations<'pl> {
-    fn push_addition(&mut self, offset: DocumentOffset<'pl>) {
-        if self.merge_method.useless_previous_changes() {
-            self.operations.clear();
-        }
-        self.operations.push(InnerDocOp::Addition(offset))
+    fn push_replacement(&mut self, offset: DocumentOffset<'pl>) {
+        self.operations.clear();
+        self.operations.push(InnerDocOp::Replace(offset))
+    }
+
+    fn push_update(&mut self, offset: DocumentOffset<'pl>) {
+        self.operations.push(InnerDocOp::Update(offset))
     }
 
     fn push_deletion(&mut self) {
@@ -439,16 +488,114 @@ impl<'pl> PayloadOperations<'pl> {
 
     fn append_operations(&mut self, mut operations: Vec<InnerDocOp<'pl>>) {
         debug_assert!(!operations.is_empty());
-        if self.merge_method.useless_previous_changes() {
+        if matches!(operations.first(), Some(InnerDocOp::Deletion | InnerDocOp::Replace(_))) {
             self.operations.clear();
         }
         self.operations.append(&mut operations);
     }
+
+    /// Returns only the most recent version of a document based on the updates from the payloads.
+    ///
+    /// This function is only meant to be used when doing a replacement and not an update.
+    fn merge<'doc>(
+        &self,
+        external_doc: &'doc str,
+        doc_alloc: &'doc Bump,
+    ) -> Result<Option<DocumentChange<'doc>>>
+    where
+        'pl: 'doc,
+    {
+        match self.operations.last() {
+            Some(InnerDocOp::Replace(DocumentOffset { content })) => {
+                let document = serde_json::from_slice(content).unwrap();
+                let document =
+                    RawMap::from_raw_value_and_hasher(document, FxBuildHasher, doc_alloc)
+                        .map_err(UserError::SerdeJson)?;
+
+                if self.is_new {
+                    Ok(Some(DocumentChange::Insertion(Insertion::create(
+                        self.docid,
+                        external_doc,
+                        Versions::single(document),
+                    ))))
+                } else {
+                    Ok(Some(DocumentChange::Update(Update::create(
+                        self.docid,
+                        external_doc,
+                        Versions::single(document),
+                        true,
+                    ))))
+                }
+            }
+            Some(InnerDocOp::Update(_)) => {
+                // Search the first operation that is a tombstone which resets the document.
+                let last_tombstone = self
+                    .operations
+                    .iter()
+                    .rposition(|op| matches!(op, InnerDocOp::Deletion | InnerDocOp::Replace(_)));
+
+                // Track when we must ignore previous document versions from the rtxn.
+                let from_scratch = last_tombstone.is_some();
+
+                // We ignore deletion and keep the replacement to create the appropriate versions.
+                let operations = match last_tombstone {
+                    Some(i) => match self.operations[i] {
+                        InnerDocOp::Deletion => &self.operations[i + 1..],
+                        InnerDocOp::Replace(_) => &self.operations[i..],
+                        InnerDocOp::Update(_) => unreachable!("Found a non-tombstone operation"),
+                    },
+                    None => &self.operations[..],
+                };
+
+                // We collect the versions to generate the appropriate document.
+                let versions = operations.iter().map(|operation| {
+                    let DocumentOffset { content } = match operation {
+                        InnerDocOp::Replace(offset) | InnerDocOp::Update(offset) => offset,
+                        InnerDocOp::Deletion => unreachable!("Deletion in document operations"),
+                    };
+
+                    let document = serde_json::from_slice(content).unwrap();
+                    let document =
+                        RawMap::from_raw_value_and_hasher(document, FxBuildHasher, doc_alloc)
+                            .map_err(UserError::SerdeJson)?;
+
+                    Ok(document)
+                });
+
+                let Some(versions) = Versions::multiple(versions)? else { return Ok(None) };
+
+                if self.is_new {
+                    Ok(Some(DocumentChange::Insertion(Insertion::create(
+                        self.docid,
+                        external_doc,
+                        versions,
+                    ))))
+                } else {
+                    Ok(Some(DocumentChange::Update(Update::create(
+                        self.docid,
+                        external_doc,
+                        versions,
+                        from_scratch,
+                    ))))
+                }
+            }
+            Some(InnerDocOp::Deletion) => {
+                return if self.is_new {
+                    Ok(None)
+                } else {
+                    let deletion = Deletion::create(self.docid, external_doc);
+                    Ok(Some(DocumentChange::Deletion(deletion)))
+                };
+            }
+            None => unreachable!("We must not have an empty set of operations on a document"),
+        }
+    }
 }
 
 #[derive(Clone)]
 pub enum InnerDocOp<'pl> {
-    Addition(DocumentOffset<'pl>),
+    Replace(DocumentOffset<'pl>),
+    Update(DocumentOffset<'pl>),
     Deletion,
 }
 
@@ -460,231 +607,14 @@ pub struct DocumentOffset<'pl> {
     pub content: &'pl [u8],
 }
 
-trait MergeChanges {
-    /// Whether the payloads in the list of operations are useless or not.
-    fn useless_previous_changes(&self) -> bool;
-
-    /// Returns a key that is used to order the payloads the right way.
-    fn sort_key(&self, docops: &[InnerDocOp]) -> usize;
-
-    fn merge<'doc>(
-        &self,
-        docid: DocumentId,
-        external_docid: &'doc str,
-        is_new: bool,
-        doc_alloc: &'doc Bump,
-        operations: &'doc [InnerDocOp],
-    ) -> Result<Option<DocumentChange<'doc>>>;
-}
-
-#[derive(Debug, Clone, Copy)]
-enum MergeMethod {
-    ForReplacement(MergeDocumentForReplacement),
-    ForUpdates(MergeDocumentForUpdates),
-}
-
-impl MergeChanges for MergeMethod {
-    fn useless_previous_changes(&self) -> bool {
-        match self {
-            MergeMethod::ForReplacement(merge) => merge.useless_previous_changes(),
-            MergeMethod::ForUpdates(merge) => merge.useless_previous_changes(),
-        }
-    }
-
-    fn sort_key(&self, docops: &[InnerDocOp]) -> usize {
-        match self {
-            MergeMethod::ForReplacement(merge) => merge.sort_key(docops),
-            MergeMethod::ForUpdates(merge) => merge.sort_key(docops),
-        }
-    }
-
-    fn merge<'doc>(
-        &self,
-        docid: DocumentId,
-        external_docid: &'doc str,
-        is_new: bool,
-        doc_alloc: &'doc Bump,
-        operations: &'doc [InnerDocOp],
-    ) -> Result<Option<DocumentChange<'doc>>> {
-        match self {
-            MergeMethod::ForReplacement(merge) => {
-                merge.merge(docid, external_docid, is_new, doc_alloc, operations)
-            }
-            MergeMethod::ForUpdates(merge) => {
-                merge.merge(docid, external_docid, is_new, doc_alloc, operations)
-            }
-        }
-    }
-}
-
-impl From<IndexDocumentsMethod> for MergeMethod {
-    fn from(method: IndexDocumentsMethod) -> Self {
-        match method {
-            IndexDocumentsMethod::ReplaceDocuments => {
-                MergeMethod::ForReplacement(MergeDocumentForReplacement)
-            }
-            IndexDocumentsMethod::UpdateDocuments => {
-                MergeMethod::ForUpdates(MergeDocumentForUpdates)
-            }
-        }
-    }
-}
-
-#[derive(Debug, Clone, Copy)]
-struct MergeDocumentForReplacement;
-
-impl MergeChanges for MergeDocumentForReplacement {
-    fn useless_previous_changes(&self) -> bool {
-        true
-    }
-
-    /// Reorders to read only the last change.
-    fn sort_key(&self, docops: &[InnerDocOp]) -> usize {
-        let f = |ido: &_| match ido {
-            InnerDocOp::Addition(add) => Some(add.content.as_ptr() as usize),
-            InnerDocOp::Deletion => None,
-        };
-        docops.iter().rev().find_map(f).unwrap_or(0)
-    }
-
-    /// Returns only the most recent version of a document based on the updates from the payloads.
-    ///
-    /// This function is only meant to be used when doing a replacement and not an update.
-    fn merge<'doc>(
-        &self,
-        docid: DocumentId,
-        external_doc: &'doc str,
-        is_new: bool,
-        doc_alloc: &'doc Bump,
-        operations: &'doc [InnerDocOp],
-    ) -> Result<Option<DocumentChange<'doc>>> {
-        match operations.last() {
-            Some(InnerDocOp::Addition(DocumentOffset { content })) => {
-                let document = serde_json::from_slice(content).unwrap();
-                let document =
-                    RawMap::from_raw_value_and_hasher(document, FxBuildHasher, doc_alloc)
-                        .map_err(UserError::SerdeJson)?;
-
-                if is_new {
-                    Ok(Some(DocumentChange::Insertion(Insertion::create(
-                        docid,
-                        external_doc,
-                        Versions::single(document),
-                    ))))
-                } else {
-                    Ok(Some(DocumentChange::Update(Update::create(
-                        docid,
-                        external_doc,
-                        Versions::single(document),
-                        true,
-                    ))))
-                }
-            }
-            Some(InnerDocOp::Deletion) => {
-                return if is_new {
-                    Ok(None)
-                } else {
-                    let deletion = Deletion::create(docid, external_doc);
-                    Ok(Some(DocumentChange::Deletion(deletion)))
-                };
-            }
-            None => unreachable!("We must not have empty set of operations on a document"),
-        }
-    }
-}
-
-#[derive(Debug, Clone, Copy)]
-struct MergeDocumentForUpdates;
-
-impl MergeChanges for MergeDocumentForUpdates {
-    fn useless_previous_changes(&self) -> bool {
-        false
-    }
-
-    /// Reorders to read the first changes first so that it's faster to read the first one and then the rest.
-    fn sort_key(&self, docops: &[InnerDocOp]) -> usize {
-        let f = |ido: &_| match ido {
-            InnerDocOp::Addition(add) => Some(add.content.as_ptr() as usize),
-            InnerDocOp::Deletion => None,
-        };
-        docops.iter().find_map(f).unwrap_or(0)
-    }
-
-    /// Reads the previous version of a document from the database, the new versions
-    /// in the grenad update files and merges them to generate a new boxed obkv.
-    ///
-    /// This function is only meant to be used when doing an update and not a replacement.
-    fn merge<'doc>(
-        &self,
-        docid: DocumentId,
-        external_docid: &'doc str,
-        is_new: bool,
-        doc_alloc: &'doc Bump,
-        operations: &'doc [InnerDocOp],
-    ) -> Result<Option<DocumentChange<'doc>>> {
-        if operations.is_empty() {
-            unreachable!("We must not have empty set of operations on a document");
-        }
-
-        let last_deletion = operations.iter().rposition(|op| matches!(op, InnerDocOp::Deletion));
-        let operations = &operations[last_deletion.map_or(0, |i| i + 1)..];
-
-        let has_deletion = last_deletion.is_some();
-
-        if operations.is_empty() {
-            return if is_new {
-                Ok(None)
-            } else {
-                let deletion = Deletion::create(docid, external_docid);
-                Ok(Some(DocumentChange::Deletion(deletion)))
-            };
-        }
-
-        let versions = match operations {
-            [single] => {
-                let DocumentOffset { content } = match single {
-                    InnerDocOp::Addition(offset) => offset,
-                    InnerDocOp::Deletion => {
-                        unreachable!("Deletion in document operations")
-                    }
-                };
-                let document = serde_json::from_slice(content).unwrap();
-                let document =
-                    RawMap::from_raw_value_and_hasher(document, FxBuildHasher, doc_alloc)
-                        .map_err(UserError::SerdeJson)?;
-
-                Some(Versions::single(document))
-            }
-            operations => {
-                let versions = operations.iter().map(|operation| {
-                    let DocumentOffset { content } = match operation {
-                        InnerDocOp::Addition(offset) => offset,
-                        InnerDocOp::Deletion => {
-                            unreachable!("Deletion in document operations")
-                        }
-                    };
-
-                    let document = serde_json::from_slice(content).unwrap();
-                    let document =
-                        RawMap::from_raw_value_and_hasher(document, FxBuildHasher, doc_alloc)
-                            .map_err(UserError::SerdeJson)?;
-                    Ok(document)
-                });
-                Versions::multiple(versions)?
-            }
-        };
-
-        let Some(versions) = versions else { return Ok(None) };
-
-        if is_new {
-            Ok(Some(DocumentChange::Insertion(Insertion::create(docid, external_docid, versions))))
-        } else {
-            Ok(Some(DocumentChange::Update(Update::create(
-                docid,
-                external_docid,
-                versions,
-                has_deletion,
-            ))))
-        }
-    }
+/// Returns the first pointer of the first change in a document.
+///
+/// This is used to sort the documents in update file content order
+/// and read the update file in order to largely speed up the indexation.
+pub fn first_update_pointer(docops: &[InnerDocOp]) -> Option<usize> {
+    docops.iter().find_map(|ido: &_| match ido {
+        InnerDocOp::Replace(replace) => Some(replace.content.as_ptr() as usize),
+        InnerDocOp::Update(update) => Some(update.content.as_ptr() as usize),
+        InnerDocOp::Deletion => None,
+    })
 }

crates/milli/tests/search/facet_distribution.rs

@@ -5,7 +5,7 @@ use maplit::hashset;
 use milli::documents::mmap_from_objects;
 use milli::progress::Progress;
 use milli::update::new::indexer;
-use milli::update::{IndexDocumentsMethod, IndexerConfig, Settings};
+use milli::update::{IndexerConfig, Settings};
 use milli::vector::EmbeddingConfigs;
 use milli::{FacetDistribution, Index, Object, OrderBy};
 use serde_json::{from_value, json};
@@ -36,7 +36,7 @@ fn test_facet_distribution_with_no_facet_values() {
     let mut new_fields_ids_map = db_fields_ids_map.clone();
 
     let embedders = EmbeddingConfigs::default();
-    let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+    let mut indexer = indexer::DocumentOperation::new();
 
     let doc1: Object = from_value(
         json!({ "id": 123, "title": "What a week, hu...", "genres": [], "tags": ["blue"] }),
@@ -47,7 +47,7 @@ fn test_facet_distribution_with_no_facet_values() {
     let documents = mmap_from_objects(vec![doc1, doc2]);
 
     // index documents
-    indexer.add_documents(&documents).unwrap();
+    indexer.replace_documents(&documents).unwrap();
 
     let indexer_alloc = Bump::new();
     let (document_changes, _operation_stats, primary_key) = indexer

crates/milli/tests/search/mod.rs

@@ -9,7 +9,7 @@ use heed::EnvOpenOptions;
 use maplit::{btreemap, hashset};
 use milli::progress::Progress;
 use milli::update::new::indexer;
-use milli::update::{IndexDocumentsMethod, IndexerConfig, Settings};
+use milli::update::{IndexerConfig, Settings};
 use milli::vector::EmbeddingConfigs;
 use milli::{AscDesc, Criterion, DocumentId, Index, Member, TermsMatchingStrategy};
 use serde::{Deserialize, Deserializer};
@@ -72,7 +72,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
     let mut new_fields_ids_map = db_fields_ids_map.clone();
 
     let embedders = EmbeddingConfigs::default();
-    let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+    let mut indexer = indexer::DocumentOperation::new();
 
     let mut file = tempfile::tempfile().unwrap();
     file.write_all(CONTENT.as_bytes()).unwrap();
@@ -80,7 +80,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
     let payload = unsafe { memmap2::Mmap::map(&file).unwrap() };
 
     // index documents
-    indexer.add_documents(&payload).unwrap();
+    indexer.replace_documents(&payload).unwrap();
 
     let indexer_alloc = Bump::new();
     let (document_changes, operation_stats, primary_key) = indexer

crates/milli/tests/search/query_criteria.rs

@@ -7,7 +7,7 @@ use itertools::Itertools;
 use maplit::hashset;
 use milli::progress::Progress;
 use milli::update::new::indexer;
-use milli::update::{IndexDocumentsMethod, IndexerConfig, Settings};
+use milli::update::{IndexerConfig, Settings};
 use milli::vector::EmbeddingConfigs;
 use milli::{AscDesc, Criterion, Index, Member, Search, SearchResult, TermsMatchingStrategy};
 use rand::Rng;
@@ -288,7 +288,7 @@ fn criteria_ascdesc() {
     let mut new_fields_ids_map = db_fields_ids_map.clone();
 
     let embedders = EmbeddingConfigs::default();
-    let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+    let mut indexer = indexer::DocumentOperation::new();
 
     let mut file = tempfile::tempfile().unwrap();
     (0..ASC_DESC_CANDIDATES_THRESHOLD + 1).for_each(|_| {
@@ -318,7 +318,7 @@ fn criteria_ascdesc() {
     file.sync_all().unwrap();
 
     let payload = unsafe { memmap2::Mmap::map(&file).unwrap() };
-    indexer.add_documents(&payload).unwrap();
+    indexer.replace_documents(&payload).unwrap();
     let (document_changes, _operation_stats, primary_key) = indexer
         .into_changes(
             &indexer_alloc,

crates/milli/tests/search/typo_tolerance.rs

@@ -5,7 +5,7 @@ use heed::EnvOpenOptions;
 use milli::documents::mmap_from_objects;
 use milli::progress::Progress;
 use milli::update::new::indexer;
-use milli::update::{IndexDocumentsMethod, IndexerConfig, Settings};
+use milli::update::{IndexerConfig, Settings};
 use milli::vector::EmbeddingConfigs;
 use milli::{Criterion, Index, Object, Search, TermsMatchingStrategy};
 use serde_json::from_value;
@@ -123,9 +123,9 @@ fn test_typo_disabled_on_word() {
     let db_fields_ids_map = index.fields_ids_map(&rtxn).unwrap();
     let mut new_fields_ids_map = db_fields_ids_map.clone();
     let embedders = EmbeddingConfigs::default();
-    let mut indexer = indexer::DocumentOperation::new(IndexDocumentsMethod::ReplaceDocuments);
+    let mut indexer = indexer::DocumentOperation::new();
 
-    indexer.add_documents(&documents).unwrap();
+    indexer.replace_documents(&documents).unwrap();
 
     let indexer_alloc = Bump::new();
     let (document_changes, _operation_stats, primary_key) = indexer

crates/xtask/src/main.rs

@@ -6,8 +6,8 @@ use xtask::bench::BenchDeriveArgs;
 /// List features available in the workspace
 #[derive(Parser, Debug)]
 struct ListFeaturesDeriveArgs {
-    /// Feature to exclude from the list. Repeat the argument to exclude multiple features
-    #[arg(short, long)]
+    /// Feature to exclude from the list. Use a comma to separate multiple features.
+    #[arg(short, long, value_delimiter = ',')]
     exclude_feature: Vec<String>,
 }