panic on serde json

5276: Fix the stuck indexation due to the internal BBQueue capacity r=curquiza a=Kerollmops

Fixes https://github.com/meilisearch/meilisearch/issues/5277. Reduce the maximum reserve grant in the BBQueue so we are never stuck.

Co-authored-by: Kerollmops <clement@meilisearch.com>
Co-authored-by: Louis Dureuil <louis@meilisearch.com>
Co-authored-by: Clément Renault <clement@meilisearch.com>

Cargo.lock

@@ -489,9 +489,14 @@ version = "0.22.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "72b3254f16251a8381aa12e40e3c4d2f0199f8c6508fbecb9d91f575e0fbb8c6"
 
+[[package]]
+name = "bbqueue"
+version = "0.5.1"
+source = "git+https://github.com/meilisearch/bbqueue#cbb87cc707b5af415ef203bdaf2443e06ba0d6d4"
+
 [[package]]
 name = "benchmarks"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "anyhow",
  "bumpalo",
@@ -684,7 +689,7 @@ dependencies = [
 
 [[package]]
 name = "build-info"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "anyhow",
  "time",
@@ -701,6 +706,20 @@ dependencies = [
  "serde",
 ]
 
+[[package]]
+name = "bumparaw-collections"
+version = "0.1.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4ce682bdc86c2e25ef5cd95881d9d6a1902214eddf74cf9ffea88fe1464377e8"
+dependencies = [
+ "allocator-api2",
+ "bitpacking",
+ "bumpalo",
+ "hashbrown 0.15.1",
+ "serde",
+ "serde_json",
+]
+
 [[package]]
 name = "byte-unit"
 version = "5.1.4"
@@ -1246,19 +1265,6 @@ dependencies = [
  "itertools 0.10.5",
 ]
 
-[[package]]
-name = "crossbeam"
-version = "0.8.4"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1137cd7e7fc0fb5d3c5a8678be38ec56e819125d8d7907411fe24ccb943faca8"
-dependencies = [
- "crossbeam-channel",
- "crossbeam-deque",
- "crossbeam-epoch",
- "crossbeam-queue",
- "crossbeam-utils",
-]
-
 [[package]]
 name = "crossbeam-channel"
 version = "0.5.13"
@@ -1658,7 +1664,7 @@ dependencies = [
 
 [[package]]
 name = "dump"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "anyhow",
  "big_s",
@@ -1870,7 +1876,7 @@ checksum = "486f806e73c5707928240ddc295403b1b93c96a02038563881c4a2fd84b81ac4"
 
 [[package]]
 name = "file-store"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "tempfile",
  "thiserror",
@@ -1892,7 +1898,7 @@ dependencies = [
 
 [[package]]
 name = "filter-parser"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "insta",
  "nom",
@@ -1912,12 +1918,21 @@ dependencies = [
 
 [[package]]
 name = "flatten-serde-json"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "criterion",
  "serde_json",
 ]
 
+[[package]]
+name = "flume"
+version = "0.11.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "da0e4dd2a88388a1f4ccc7c9ce104604dab68d9f408dc34cd45823d5a9069095"
+dependencies = [
+ "spin",
+]
+
 [[package]]
 name = "fnv"
 version = "1.0.7"
@@ -2042,7 +2057,7 @@ dependencies = [
 
 [[package]]
 name = "fuzzers"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "arbitrary",
  "bumpalo",
@@ -2609,14 +2624,16 @@ checksum = "206ca75c9c03ba3d4ace2460e57b189f39f43de612c2f85836e65c929701bb2d"
 
 [[package]]
 name = "index-scheduler"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "anyhow",
  "arroy 0.5.0 (registry+https://github.com/rust-lang/crates.io-index)",
  "big_s",
  "bincode",
  "bumpalo",
- "crossbeam",
+ "bumparaw-collections",
+ "convert_case 0.6.0",
+ "crossbeam-channel",
  "csv",
  "derive_builder 0.20.0",
  "dump",
@@ -2630,7 +2647,6 @@ dependencies = [
  "meilisearch-types",
  "memmap2",
  "page_size",
- "raw-collections",
  "rayon",
  "roaring",
  "serde",
@@ -2646,12 +2662,12 @@ dependencies = [
 
 [[package]]
 name = "indexmap"
-version = "2.2.6"
+version = "2.7.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "168fb715dda47215e360912c096649d23d58bf392ac62f73919e831745e40f26"
+checksum = "62f822373a4fe84d4bb149bf54e584a7f4abec90e072ed49cda0edea5b95471f"
 dependencies = [
  "equivalent",
- "hashbrown 0.14.3",
+ "hashbrown 0.15.1",
  "serde",
 ]
 
@@ -2806,7 +2822,7 @@ dependencies = [
 
 [[package]]
 name = "json-depth-checker"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "criterion",
  "serde_json",
@@ -3425,7 +3441,7 @@ checksum = "490cc448043f947bae3cbee9c203358d62dbee0db12107a74be5c30ccfd09771"
 
 [[package]]
 name = "meili-snap"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "insta",
  "md5",
@@ -3434,7 +3450,7 @@ dependencies = [
 
 [[package]]
 name = "meilisearch"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "actix-cors",
  "actix-http",
@@ -3524,7 +3540,7 @@ dependencies = [
 
 [[package]]
 name = "meilisearch-auth"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "base64 0.22.1",
  "enum-iterator",
@@ -3543,11 +3559,12 @@ dependencies = [
 
 [[package]]
 name = "meilisearch-types"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "actix-web",
  "anyhow",
  "bumpalo",
+ "bumparaw-collections",
  "convert_case 0.6.0",
  "csv",
  "deserr",
@@ -3560,8 +3577,8 @@ dependencies = [
  "meili-snap",
  "memmap2",
  "milli",
- "raw-collections",
  "roaring",
+ "rustc-hash 2.1.0",
  "serde",
  "serde-cs",
  "serde_json",
@@ -3575,16 +3592,19 @@ dependencies = [
 
 [[package]]
 name = "meilitool"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "anyhow",
  "arroy 0.5.0 (git+https://github.com/meilisearch/arroy/?tag=DO-NOT-DELETE-upgrade-v04-to-v05)",
  "clap",
  "dump",
  "file-store",
+ "indexmap",
  "meilisearch-auth",
  "meilisearch-types",
  "serde",
+ "serde_json",
+ "tempfile",
  "time",
  "uuid",
 ]
@@ -3607,15 +3627,17 @@ dependencies = [
 
 [[package]]
 name = "milli"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "allocator-api2",
  "arroy 0.5.0 (registry+https://github.com/rust-lang/crates.io-index)",
+ "bbqueue",
  "big_s",
  "bimap",
  "bincode",
  "bstr",
  "bumpalo",
+ "bumparaw-collections",
  "bytemuck",
  "byteorder",
  "candle-core",
@@ -3630,6 +3652,7 @@ dependencies = [
  "enum-iterator",
  "filter-parser",
  "flatten-serde-json",
+ "flume",
  "fst",
  "fxhash",
  "geoutils",
@@ -3653,13 +3676,12 @@ dependencies = [
  "once_cell",
  "ordered-float",
  "rand",
- "raw-collections",
  "rayon",
  "rayon-par-bridge",
  "rhai",
  "roaring",
  "rstar",
- "rustc-hash 2.0.0",
+ "rustc-hash 2.1.0",
  "serde",
  "serde_json",
  "slice-group-by",
@@ -4061,7 +4083,7 @@ checksum = "e3148f5046208a5d56bcfc03053e3ca6334e51da8dfb19b6cdc8b306fae3283e"
 
 [[package]]
 name = "permissive-json-pointer"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "big_s",
  "serde_json",
@@ -4408,7 +4430,7 @@ dependencies = [
  "bytes",
  "rand",
  "ring",
- "rustc-hash 2.0.0",
+ "rustc-hash 2.1.0",
  "rustls",
  "slab",
  "thiserror",
@@ -4484,19 +4506,6 @@ dependencies = [
  "rand",
 ]
 
-[[package]]
-name = "raw-collections"
-version = "0.1.0"
-source = "git+https://github.com/meilisearch/raw-collections.git#15e5d7bdebc0c149b2a28b2454f307c717d07f8a"
-dependencies = [
- "allocator-api2",
- "bitpacking",
- "bumpalo",
- "hashbrown 0.15.1",
- "serde",
- "serde_json",
-]
-
 [[package]]
 name = "raw-cpuid"
 version = "10.7.0"
@@ -4743,8 +4752,9 @@ dependencies = [
 
 [[package]]
 name = "roaring"
-version = "0.10.6"
-source = "git+https://github.com/RoaringBitmap/roaring-rs?branch=clone-iter-slice#8ff028e484fb6192a0acf5a669eaf18c30cada6e"
+version = "0.10.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f81dc953b2244ddd5e7860cb0bb2a790494b898ef321d4aff8e260efab60cc88"
 dependencies = [
  "bytemuck",
  "byteorder",
@@ -4793,9 +4803,9 @@ checksum = "08d43f7aa6b08d49f382cde6a7982047c3426db949b1424bc4b7ec9ae12c6ce2"
 
 [[package]]
 name = "rustc-hash"
-version = "2.0.0"
+version = "2.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "583034fd73374156e66797ed8e5b0d5690409c9226b22d87cb7f19821c05d152"
+checksum = "c7fb8039b3032c191086b10f11f319a6e99e1e82889c5cc6046f515c9db1d497"
 
 [[package]]
 name = "rustc_version"
@@ -4964,9 +4974,9 @@ dependencies = [
 
 [[package]]
 name = "serde_json"
-version = "1.0.132"
+version = "1.0.133"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d726bfaff4b320266d395898905d0eba0345aae23b54aee3a737e260fd46db03"
+checksum = "c7fceb2473b9166b2294ef05efcb65a3db80803f0b03ef86a5fc88a2b85ee377"
 dependencies = [
  "indexmap",
  "itoa",
@@ -5186,6 +5196,9 @@ name = "spin"
 version = "0.9.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "6980e8d7511241f8acf4aebddbb1ff938df5eebe98691418c4468d0b72a96a67"
+dependencies = [
+ "lock_api",
+]
 
 [[package]]
 name = "spm_precompiled"
@@ -6473,7 +6486,7 @@ dependencies = [
 
 [[package]]
 name = "xtask"
-version = "1.12.0"
+version = "1.12.7"
 dependencies = [
  "anyhow",
  "build-info",

Cargo.toml

@@ -22,7 +22,7 @@ members = [
 ]
 
 [workspace.package]
-version = "1.12.0"
+version = "1.12.7"
 authors = [
     "Quentin de Quelen <quentin@dequelen.me>",
     "Clément Renault <clement@meilisearch.com>",
@@ -43,6 +43,3 @@ opt-level = 3
 opt-level = 3
 [profile.dev.package.roaring]
 opt-level = 3
-
-[patch.crates-io]
-roaring = { git = "https://github.com/RoaringBitmap/roaring-rs", branch = "clone-iter-slice" }

crates/benchmarks/Cargo.toml

@@ -24,7 +24,7 @@ tempfile = "3.14.0"
 criterion = { version = "0.5.1", features = ["html_reports"] }
 rand = "0.8.5"
 rand_chacha = "0.3.1"
-roaring = "0.10.6"
+roaring = "0.10.7"
 
 [build-dependencies]
 anyhow = "1.0.86"

crates/benchmarks/benches/indexing.rs

@@ -8,6 +8,7 @@ use bumpalo::Bump;
 use criterion::{criterion_group, criterion_main, Criterion};
 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::vector::EmbeddingConfigs;
@@ -16,6 +17,7 @@ use rand::seq::SliceRandom;
 use rand_chacha::rand_core::SeedableRng;
 use roaring::RoaringBitmap;
 
+#[cfg(not(windows))]
 #[global_allocator]
 static ALLOC: mimalloc::MiMalloc = mimalloc::MiMalloc;
 
@@ -150,13 +152,14 @@ fn indexing_songs_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -164,7 +167,7 @@ fn indexing_songs_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -216,13 +219,14 @@ fn reindexing_songs_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -230,7 +234,7 @@ fn reindexing_songs_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -260,13 +264,14 @@ fn reindexing_songs_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -274,7 +279,7 @@ fn reindexing_songs_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -328,13 +333,14 @@ fn deleting_songs_in_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -342,7 +348,7 @@ fn deleting_songs_in_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -404,13 +410,14 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -418,7 +425,7 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -448,13 +455,14 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -462,7 +470,7 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -488,13 +496,14 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -502,7 +511,7 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -555,13 +564,14 @@ fn indexing_songs_without_faceted_numbers(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -569,7 +579,7 @@ fn indexing_songs_without_faceted_numbers(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -621,13 +631,14 @@ fn indexing_songs_without_faceted_fields(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -635,7 +646,7 @@ fn indexing_songs_without_faceted_fields(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -687,13 +698,14 @@ fn indexing_wiki(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -701,7 +713,7 @@ fn indexing_wiki(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -752,13 +764,14 @@ fn reindexing_wiki(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -766,7 +779,7 @@ fn reindexing_wiki(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -796,13 +809,14 @@ fn reindexing_wiki(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -810,7 +824,7 @@ fn reindexing_wiki(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -863,13 +877,14 @@ fn deleting_wiki_in_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -877,7 +892,7 @@ fn deleting_wiki_in_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -939,13 +954,14 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -953,7 +969,7 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -984,13 +1000,14 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -998,7 +1015,7 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -1025,13 +1042,14 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1039,7 +1057,7 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -1091,13 +1109,14 @@ fn indexing_movies_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1105,7 +1124,7 @@ fn indexing_movies_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -1156,13 +1175,14 @@ fn reindexing_movies_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1170,7 +1190,7 @@ fn reindexing_movies_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -1200,13 +1220,14 @@ fn reindexing_movies_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1214,7 +1235,7 @@ fn reindexing_movies_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -1267,13 +1288,14 @@ fn deleting_movies_in_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1281,7 +1303,7 @@ fn deleting_movies_in_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -1321,6 +1343,7 @@ fn delete_documents_from_ids(index: Index, document_ids_to_delete: Vec<RoaringBi
         indexer::index(
             &mut wtxn,
             &index,
+            &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
             config.grenad_parameters(),
             &db_fields_ids_map,
             new_fields_ids_map,
@@ -1328,7 +1351,7 @@ fn delete_documents_from_ids(index: Index, document_ids_to_delete: Vec<RoaringBi
             &document_changes,
             EmbeddingConfigs::default(),
             &|| false,
-            &|_| (),
+            &Progress::default(),
         )
         .unwrap();
 
@@ -1378,13 +1401,14 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1392,7 +1416,7 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -1422,13 +1446,14 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1436,7 +1461,7 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -1462,13 +1487,14 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1476,7 +1502,7 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -1551,13 +1577,14 @@ fn indexing_nested_movies_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1565,7 +1592,7 @@ fn indexing_nested_movies_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -1641,13 +1668,14 @@ fn deleting_nested_movies_in_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1655,7 +1683,7 @@ fn deleting_nested_movies_in_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -1723,13 +1751,14 @@ fn indexing_nested_movies_without_faceted_fields(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1737,7 +1766,7 @@ fn indexing_nested_movies_without_faceted_fields(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -1789,13 +1818,14 @@ fn indexing_geo(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1803,7 +1833,7 @@ fn indexing_geo(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -1854,13 +1884,14 @@ fn reindexing_geo(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1868,7 +1899,7 @@ fn reindexing_geo(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -1898,13 +1929,14 @@ fn reindexing_geo(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1912,7 +1944,7 @@ fn reindexing_geo(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 
@@ -1965,13 +1997,14 @@ fn deleting_geo_in_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        &|_progress| (),
+                        Progress::default(),
                     )
                     .unwrap();
 
                 indexer::index(
                     &mut wtxn,
                     &index,
+                    &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1979,7 +2012,7 @@ fn deleting_geo_in_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
                 .unwrap();
 

crates/benchmarks/benches/search_geo.rs

@@ -5,6 +5,7 @@ use criterion::{criterion_group, criterion_main};
 use milli::update::Settings;
 use utils::Conf;
 
+#[cfg(not(windows))]
 #[global_allocator]
 static ALLOC: mimalloc::MiMalloc = mimalloc::MiMalloc;
 

crates/benchmarks/benches/search_songs.rs

@@ -5,6 +5,7 @@ use criterion::{criterion_group, criterion_main};
 use milli::update::Settings;
 use utils::Conf;
 
+#[cfg(not(windows))]
 #[global_allocator]
 static ALLOC: mimalloc::MiMalloc = mimalloc::MiMalloc;
 

crates/benchmarks/benches/search_wiki.rs

@@ -5,6 +5,7 @@ use criterion::{criterion_group, criterion_main};
 use milli::update::Settings;
 use utils::Conf;
 
+#[cfg(not(windows))]
 #[global_allocator]
 static ALLOC: mimalloc::MiMalloc = mimalloc::MiMalloc;
 

crates/benchmarks/benches/utils.rs

@@ -10,6 +10,7 @@ use bumpalo::Bump;
 use criterion::BenchmarkId;
 use memmap2::Mmap;
 use milli::heed::EnvOpenOptions;
+use milli::progress::Progress;
 use milli::update::new::indexer;
 use milli::update::{IndexDocumentsMethod, IndexerConfig, Settings};
 use milli::vector::EmbeddingConfigs;
@@ -110,13 +111,14 @@ pub fn base_setup(conf: &Conf) -> Index {
             None,
             &mut new_fields_ids_map,
             &|| false,
-            &|_progress| (),
+            Progress::default(),
         )
         .unwrap();
 
     indexer::index(
         &mut wtxn,
         &index,
+        &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
         config.grenad_parameters(),
         &db_fields_ids_map,
         new_fields_ids_map,
@@ -124,7 +126,7 @@ pub fn base_setup(conf: &Conf) -> Index {
         &document_changes,
         EmbeddingConfigs::default(),
         &|| false,
-        &|_| (),
+        &Progress::default(),
     )
     .unwrap();
 

crates/dump/Cargo.toml

@@ -17,7 +17,7 @@ http = "1.1.0"
 meilisearch-types = { path = "../meilisearch-types" }
 once_cell = "1.19.0"
 regex = "1.10.5"
-roaring = { version = "0.10.6", features = ["serde"] }
+roaring = { version = "0.10.7", features = ["serde"] }
 serde = { version = "1.0.204", features = ["derive"] }
 serde_json = { version = "1.0.120", features = ["preserve_order"] }
 tar = "0.4.41"

crates/file-store/src/lib.rs

@@ -136,6 +136,14 @@ pub struct File {
 }
 
 impl File {
+    pub fn from_parts(path: PathBuf, file: Option<NamedTempFile>) -> Self {
+        Self { path, file }
+    }
+
+    pub fn into_parts(self) -> (PathBuf, Option<NamedTempFile>) {
+        (self.path, self.file)
+    }
+
     pub fn dry_file() -> Result<Self> {
         Ok(Self { path: PathBuf::new(), file: None })
     }

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

@@ -10,6 +10,7 @@ use either::Either;
 use fuzzers::Operation;
 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::vector::EmbeddingConfigs;
@@ -128,13 +129,14 @@ fn main() {
                                     None,
                                     &mut new_fields_ids_map,
                                     &|| false,
-                                    &|_progress| (),
+                                    Progress::default(),
                                 )
                                 .unwrap();
 
                             indexer::index(
                                 &mut wtxn,
                                 &index,
+                                &milli::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                                 indexer_config.grenad_parameters(),
                                 &db_fields_ids_map,
                                 new_fields_ids_map,
@@ -142,7 +144,7 @@ fn main() {
                                 &document_changes,
                                 embedders,
                                 &|| false,
-                                &|_| (),
+                                &Progress::default(),
                             )
                             .unwrap();
 

crates/index-scheduler/Cargo.toml

@@ -13,6 +13,9 @@ license.workspace = true
 [dependencies]
 anyhow = "1.0.86"
 bincode = "1.3.3"
+bumpalo = "3.16.0"
+bumparaw-collections = "0.1.2"
+convert_case = "0.6.0"
 csv = "1.3.0"
 derive_builder = "0.20.0"
 dump = { path = "../dump" }
@@ -21,16 +24,15 @@ file-store = { path = "../file-store" }
 flate2 = "1.0.30"
 meilisearch-auth = { path = "../meilisearch-auth" }
 meilisearch-types = { path = "../meilisearch-types" }
+memmap2 = "0.9.4"
 page_size = "0.6.0"
-raw-collections = { git = "https://github.com/meilisearch/raw-collections.git", version = "0.1.0" }
 rayon = "1.10.0"
-roaring = { version = "0.10.6", features = ["serde"] }
+roaring = { version = "0.10.7", features = ["serde"] }
 serde = { version = "1.0.204", features = ["derive"] }
 serde_json = { version = "1.0.120", features = ["preserve_order"] }
 synchronoise = "1.0.1"
 tempfile = "3.10.1"
 thiserror = "1.0.61"
-memmap2 = "0.9.4"
 time = { version = "0.3.36", features = [
     "serde-well-known",
     "formatting",
@@ -40,12 +42,11 @@ time = { version = "0.3.36", features = [
 tracing = "0.1.40"
 ureq = "2.10.0"
 uuid = { version = "1.10.0", features = ["serde", "v4"] }
-bumpalo = "3.16.0"
 
 [dev-dependencies]
 arroy = "0.5.0"
 big_s = "1.0.2"
-crossbeam = "0.8.4"
+crossbeam-channel = "0.5.13"
 insta = { version = "1.39.0", features = ["json", "redactions"] }
 maplit = "1.0.2"
 meili-snap = { path = "../meili-snap" }

crates/index-scheduler/src/autobatcher.rs

@@ -115,13 +115,6 @@ pub enum BatchKind {
         allow_index_creation: bool,
         settings_ids: Vec<TaskId>,
     },
-    SettingsAndDocumentOperation {
-        settings_ids: Vec<TaskId>,
-        method: IndexDocumentsMethod,
-        allow_index_creation: bool,
-        primary_key: Option<String>,
-        operation_ids: Vec<TaskId>,
-    },
     Settings {
         allow_index_creation: bool,
         settings_ids: Vec<TaskId>,
@@ -146,7 +139,6 @@ impl BatchKind {
         match self {
             BatchKind::DocumentOperation { allow_index_creation, .. }
             | BatchKind::ClearAndSettings { allow_index_creation, .. }
-            | BatchKind::SettingsAndDocumentOperation { allow_index_creation, .. }
             | BatchKind::Settings { allow_index_creation, .. } => Some(*allow_index_creation),
             _ => None,
         }
@@ -154,10 +146,7 @@ impl BatchKind {
 
     fn primary_key(&self) -> Option<Option<&str>> {
         match self {
-            BatchKind::DocumentOperation { primary_key, .. }
-            | BatchKind::SettingsAndDocumentOperation { primary_key, .. } => {
-                Some(primary_key.as_deref())
-            }
+            BatchKind::DocumentOperation { primary_key, .. } => Some(primary_key.as_deref()),
             _ => None,
         }
     }
@@ -275,8 +264,7 @@ impl BatchKind {
                 Break(BatchKind::IndexDeletion { ids })
             }
             (
-                BatchKind::ClearAndSettings { settings_ids: mut ids, allow_index_creation: _, mut other }
-                | BatchKind::SettingsAndDocumentOperation { operation_ids: mut ids, method: _, allow_index_creation: _, primary_key: _, settings_ids: mut other },
+                BatchKind::ClearAndSettings { settings_ids: mut ids, allow_index_creation: _, mut other },
                 K::IndexDeletion,
             ) => {
                 ids.push(id);
@@ -356,15 +344,9 @@ impl BatchKind {
             ) => Break(this),
 
             (
-                BatchKind::DocumentOperation { method, allow_index_creation, primary_key, operation_ids },
+                this @ BatchKind::DocumentOperation { .. },
                 K::Settings { .. },
-            ) => Continue(BatchKind::SettingsAndDocumentOperation {
-                settings_ids: vec![id],
-                method,
-                allow_index_creation,
-                primary_key,
-                operation_ids,
-            }),
+            ) => Break(this),
 
             (BatchKind::DocumentDeletion { mut deletion_ids, includes_by_filter: _ }, K::DocumentClear) => {
                 deletion_ids.push(id);
@@ -477,63 +459,7 @@ impl BatchKind {
                     allow_index_creation,
                 })
             }
-            (
-                BatchKind::SettingsAndDocumentOperation { settings_ids, method: _, mut operation_ids, allow_index_creation, primary_key: _ },
-                K::DocumentClear,
-            ) => {
-                operation_ids.push(id);
-                Continue(BatchKind::ClearAndSettings {
-                    settings_ids,
-                    other: operation_ids,
-                    allow_index_creation,
-                })
-            }
 
-            (
-                BatchKind::SettingsAndDocumentOperation { settings_ids, method: ReplaceDocuments, mut operation_ids, allow_index_creation, primary_key: _},
-                K::DocumentImport { method: ReplaceDocuments, primary_key: pk2, .. },
-            ) => {
-                operation_ids.push(id);
-                Continue(BatchKind::SettingsAndDocumentOperation {
-                    settings_ids,
-                    method: ReplaceDocuments,
-                    allow_index_creation,
-                        primary_key: pk2,
-                    operation_ids,
-                })
-            }
-            (
-                BatchKind::SettingsAndDocumentOperation { settings_ids, method: UpdateDocuments, allow_index_creation, primary_key: _, mut operation_ids },
-                K::DocumentImport { method: UpdateDocuments, primary_key: pk2, .. },
-            ) => {
-                operation_ids.push(id);
-                Continue(BatchKind::SettingsAndDocumentOperation {
-                    settings_ids,
-                    method: UpdateDocuments,
-                    allow_index_creation,
-                    primary_key: pk2,
-                    operation_ids,
-                })
-            }
-            // But we can't batch a settings and a doc op with another doc op
-            // this MUST be AFTER the two previous branch
-            (
-                this @ BatchKind::SettingsAndDocumentOperation { .. },
-                K::DocumentDeletion { .. } | K::DocumentImport { .. },
-            ) => Break(this),
-            (
-                BatchKind::SettingsAndDocumentOperation { mut settings_ids, method, allow_index_creation,primary_key, operation_ids },
-                K::Settings { .. },
-            ) => {
-                settings_ids.push(id);
-                Continue(BatchKind::SettingsAndDocumentOperation {
-                    settings_ids,
-                    method,
-                    allow_index_creation,
-                    primary_key,
-                    operation_ids,
-                })
-            }
             (
                 BatchKind::IndexCreation { .. }
                 | BatchKind::IndexDeletion { .. }
@@ -808,30 +734,30 @@ mod tests {
     }
 
     #[test]
-    fn document_addition_batch_with_settings() {
+    fn document_addition_doesnt_batch_with_settings() {
         // simple case
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true)]), @"Some((SettingsAndDocumentOperation { settings_ids: [1], 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((SettingsAndDocumentOperation { settings_ids: [1], 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 { 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))");
 
         // 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((SettingsAndDocumentOperation { settings_ids: [2, 3], 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((SettingsAndDocumentOperation { settings_ids: [2, 3], 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 { method: ReplaceDocuments, 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((SettingsAndDocumentOperation { settings_ids: [1, 3], method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 2] }, true))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), doc_imp(UpdateDocuments, true, None), settings(true)]), @"Some((SettingsAndDocumentOperation { settings_ids: [1, 3], method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 2] }, true))");
+        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))");
 
-        // We ensure this kind of batch doesn't batch with forbidden operations
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), doc_imp(UpdateDocuments, true, None)]), @"Some((SettingsAndDocumentOperation { settings_ids: [1], 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((SettingsAndDocumentOperation { settings_ids: [1], 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((SettingsAndDocumentOperation { settings_ids: [1], 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((SettingsAndDocumentOperation { settings_ids: [1], 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((SettingsAndDocumentOperation { settings_ids: [1], 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((SettingsAndDocumentOperation { settings_ids: [1], 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((SettingsAndDocumentOperation { settings_ids: [1], 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((SettingsAndDocumentOperation { settings_ids: [1], 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((SettingsAndDocumentOperation { settings_ids: [1], 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((SettingsAndDocumentOperation { settings_ids: [1], method: UpdateDocuments, 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))");
     }
 
     #[test]
@@ -859,8 +785,8 @@ mod tests {
         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((ClearAndSettings { other: [0, 2], allow_index_creation: true, settings_ids: [1] }, true))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), doc_clr()]), @"Some((ClearAndSettings { other: [0, 2], allow_index_creation: true, settings_ids: [1] }, 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))");
     }
 
     #[test]
@@ -907,50 +833,6 @@ mod tests {
         debug_snapshot!(autobatch_from(false,None,  [doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, false))");
         debug_snapshot!(autobatch_from(false,None,  [settings(true), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, true))");
         debug_snapshot!(autobatch_from(false,None,  [settings(false), idx_del()]), @"Some((IndexDeletion { ids: [0, 1] }, false))");
-
-        // Then the mixed cases.
-        // The index already exists, whatever is the right of the tasks it shouldn't change the result.
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), idx_del()]), @"Some((IndexDeletion { ids: [0, 2, 1] }, true))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), idx_del()]), @"Some((IndexDeletion { ids: [0, 2, 1] }, true))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), settings(true), doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [1, 3, 0, 2] }, true))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(true), doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [1, 3, 0, 2] }, true))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments,false, None), settings(false), idx_del()]), @"Some((IndexDeletion { ids: [0, 2, 1] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None), settings(false), idx_del()]), @"Some((IndexDeletion { ids: [0, 2, 1] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments,false, None), settings(false), doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [1, 3, 0, 2] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None), settings(false), doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [1, 3, 0, 2] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments,false, None), settings(true), idx_del()]), @"Some((IndexDeletion { ids: [0, 2, 1] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None), settings(true), idx_del()]), @"Some((IndexDeletion { ids: [0, 2, 1] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments,false, None), settings(true), doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [1, 3, 0, 2] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None), settings(true), doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [1, 3, 0, 2] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments,true, None), settings(false), idx_del()]), @"Some((IndexDeletion { ids: [0, 2, 1] }, true))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(false), idx_del()]), @"Some((IndexDeletion { ids: [0, 2, 1] }, true))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments,true, None), settings(false), doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [1, 3, 0, 2] }, true))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), settings(false), doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [1, 3, 0, 2] }, true))");
-
-        // When the index doesn't exists yet it's more complicated.
-        // Either the first task we encounter create it, in which case we can create a big batch with everything.
-        debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, true, None), settings(true), idx_del()]), @"Some((IndexDeletion { ids: [0, 2, 1] }, true))");
-        debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, true, None), settings(true), idx_del()]), @"Some((IndexDeletion { ids: [0, 2, 1] }, true))");
-        debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments, true, None), settings(true), doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [1, 3, 0, 2] }, true))");
-        debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, true, None), settings(true), doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [1, 3, 0, 2] }, true))");
-        // The right of the tasks following isn't really important.
-        debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments,true, None), settings(false), idx_del()]), @"Some((IndexDeletion { ids: [0, 2, 1] }, true))");
-        debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, true, None), settings(false), idx_del()]), @"Some((IndexDeletion { ids: [0, 2, 1] }, true))");
-        debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments,true, None), settings(false), doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [1, 3, 0, 2] }, true))");
-        debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, true, None), settings(false), doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [1, 3, 0, 2] }, true))");
-        // Or, the second case; the first task doesn't create the index and thus we wants to batch it with only tasks that can't create an index.
-        // that can be a second task that don't have the right to create an index. Or anything that can't create an index like an index deletion, document deletion, document clear, etc.
-        // All theses tasks are going to throw an error `Index doesn't exist` once the batch is processed.
-        debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments,false, None), settings(false), idx_del()]), @"Some((IndexDeletion { ids: [0, 2, 1] }, false))");
-        debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, false, None), settings(false), idx_del()]), @"Some((IndexDeletion { ids: [0, 2, 1] }, false))");
-        debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments,false, None), settings(false), doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [1, 3, 0, 2] }, false))");
-        debug_snapshot!(autobatch_from(false,None,  [doc_imp(UpdateDocuments, false, None), settings(false), doc_clr(), idx_del()]), @"Some((IndexDeletion { ids: [1, 3, 0, 2] }, false))");
-        // The third and final case is when the first task doesn't create an index but is directly followed by a task creating an index. In this case we can't batch whit what
-        // follows because we first need to process the erronous batch.
-        debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments,false, None), settings(true), idx_del()]), @"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), settings(true), idx_del()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(false,None,  [doc_imp(ReplaceDocuments,false, None), settings(true), doc_clr(), idx_del()]), @"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), settings(true), doc_clr(), idx_del()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
     }
 
     #[test]
@@ -959,13 +841,13 @@ mod tests {
         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((SettingsAndDocumentOperation { settings_ids: [1], 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((SettingsAndDocumentOperation { settings_ids: [1], method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, 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(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((SettingsAndDocumentOperation { settings_ids: [1], method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
+        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))");
 
         // batch deletion and addition

crates/index-scheduler/src/error.rs

@@ -104,7 +104,7 @@ pub enum Error {
     )]
     InvalidTaskCanceledBy { canceled_by: String },
     #[error(
-        "{index_uid} 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."
+        "{index_uid} 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 400 bytes."
     )]
     InvalidIndexUid { index_uid: String },
     #[error("Task `{0}` not found.")]
@@ -122,8 +122,11 @@ pub enum Error {
     Dump(#[from] dump::Error),
     #[error(transparent)]
     Heed(#[from] heed::Error),
-    #[error(transparent)]
-    Milli(#[from] milli::Error),
+    #[error("{}", match .index_uid {
+        Some(uid) if !uid.is_empty() => format!("Index `{}`: {error}", uid),
+        _ => format!("{error}")
+    })]
+    Milli { error: milli::Error, index_uid: Option<String> },
     #[error("An unexpected crash occurred when processing the task.")]
     ProcessBatchPanicked,
     #[error(transparent)]
@@ -190,7 +193,7 @@ impl Error {
             | Error::AbortedTask
             | Error::Dump(_)
             | Error::Heed(_)
-            | Error::Milli(_)
+            | Error::Milli { .. }
             | Error::ProcessBatchPanicked
             | Error::FileStore(_)
             | Error::IoError(_)
@@ -209,6 +212,20 @@ impl Error {
     pub fn with_custom_error_code(self, code: Code) -> Self {
         Self::WithCustomErrorCode(code, Box::new(self))
     }
+
+    pub fn from_milli(err: milli::Error, index_uid: Option<String>) -> Self {
+        match err {
+            milli::Error::UserError(milli::UserError::InvalidFilter(_)) => {
+                Self::Milli { error: err, index_uid }
+                    .with_custom_error_code(Code::InvalidDocumentFilter)
+            }
+            milli::Error::UserError(milli::UserError::InvalidFilterExpression { .. }) => {
+                Self::Milli { error: err, index_uid }
+                    .with_custom_error_code(Code::InvalidDocumentFilter)
+            }
+            _ => Self::Milli { error: err, index_uid },
+        }
+    }
 }
 
 impl ErrorCode for Error {
@@ -236,7 +253,7 @@ impl ErrorCode for Error {
             // TODO: not sure of the Code to use
             Error::NoSpaceLeftInTaskQueue => Code::NoSpaceLeftOnDevice,
             Error::Dump(e) => e.error_code(),
-            Error::Milli(e) => e.error_code(),
+            Error::Milli { error, .. } => error.error_code(),
             Error::ProcessBatchPanicked => Code::Internal,
             Error::Heed(e) => e.error_code(),
             Error::HeedTransaction(e) => e.error_code(),

crates/index-scheduler/src/index_mapper/index_map.rs

@@ -1,16 +1,17 @@
 use std::collections::BTreeMap;
+use std::env::VarError;
 use std::path::Path;
+use std::str::FromStr;
 use std::time::Duration;
 
 use meilisearch_types::heed::{EnvClosingEvent, EnvFlags, EnvOpenOptions};
-use meilisearch_types::milli::Index;
+use meilisearch_types::milli::{Index, Result};
 use time::OffsetDateTime;
 use uuid::Uuid;
 
 use super::IndexStatus::{self, Available, BeingDeleted, Closing, Missing};
+use crate::clamp_to_page_size;
 use crate::lru::{InsertionOutcome, LruMap};
-use crate::{clamp_to_page_size, Result};
-
 /// Keep an internally consistent view of the open indexes in memory.
 ///
 /// This view is made of an LRU cache that will evict the least frequently used indexes when new indexes are opened.
@@ -303,7 +304,15 @@ fn create_or_open_index(
 ) -> Result<Index> {
     let mut options = EnvOpenOptions::new();
     options.map_size(clamp_to_page_size(map_size));
-    options.max_readers(1024);
+
+    let max_readers = match std::env::var("MEILI_EXPERIMENTAL_INDEX_MAX_READERS") {
+        Ok(value) => u32::from_str(&value).unwrap(),
+        Err(VarError::NotPresent) => 1024,
+        Err(VarError::NotUnicode(value)) => panic!(
+            "Invalid unicode for the `MEILI_EXPERIMENTAL_INDEX_MAX_READERS` env var: {value:?}"
+        ),
+    };
+    options.max_readers(max_readers);
     if enable_mdb_writemap {
         unsafe { options.flags(EnvFlags::WRITE_MAP) };
     }

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

@@ -5,6 +5,7 @@ use std::{fs, thread};
 
 use meilisearch_types::heed::types::{SerdeJson, Str};
 use meilisearch_types::heed::{Database, Env, RoTxn, RwTxn};
+use meilisearch_types::milli;
 use meilisearch_types::milli::update::IndexerConfig;
 use meilisearch_types::milli::{FieldDistribution, Index};
 use serde::{Deserialize, Serialize};
@@ -121,7 +122,7 @@ impl IndexStats {
     /// # Parameters
     ///
     /// - rtxn: a RO transaction for the index, obtained from `Index::read_txn()`.
-    pub fn new(index: &Index, rtxn: &RoTxn) -> Result<Self> {
+    pub fn new(index: &Index, rtxn: &RoTxn) -> milli::Result<Self> {
         Ok(IndexStats {
             number_of_documents: index.number_of_documents(rtxn)?,
             database_size: index.on_disk_size()?,
@@ -183,13 +184,18 @@ impl IndexMapper {
                 // Error if the UUIDv4 somehow already exists in the map, since it should be fresh.
                 // This is very unlikely to happen in practice.
                 // TODO: it would be better to lazily create the index. But we need an Index::open function for milli.
-                let index = self.index_map.write().unwrap().create(
-                    &uuid,
-                    &index_path,
-                    date,
-                    self.enable_mdb_writemap,
-                    self.index_base_map_size,
-                )?;
+                let index = self
+                    .index_map
+                    .write()
+                    .unwrap()
+                    .create(
+                        &uuid,
+                        &index_path,
+                        date,
+                        self.enable_mdb_writemap,
+                        self.index_base_map_size,
+                    )
+                    .map_err(|e| Error::from_milli(e, Some(uuid.to_string())))?;
 
                 wtxn.commit()?;
 
@@ -357,7 +363,9 @@ impl IndexMapper {
                     };
                     let index_path = self.base_path.join(uuid.to_string());
                     // take the lock to reopen the environment.
-                    reopen.reopen(&mut self.index_map.write().unwrap(), &index_path)?;
+                    reopen
+                        .reopen(&mut self.index_map.write().unwrap(), &index_path)
+                        .map_err(|e| Error::from_milli(e, Some(uuid.to_string())))?;
                     continue;
                 }
                 BeingDeleted => return Err(Error::IndexNotFound(name.to_string())),
@@ -372,13 +380,15 @@ impl IndexMapper {
                         Missing => {
                             let index_path = self.base_path.join(uuid.to_string());
 
-                            break index_map.create(
-                                &uuid,
-                                &index_path,
-                                None,
-                                self.enable_mdb_writemap,
-                                self.index_base_map_size,
-                            )?;
+                            break index_map
+                                .create(
+                                    &uuid,
+                                    &index_path,
+                                    None,
+                                    self.enable_mdb_writemap,
+                                    self.index_base_map_size,
+                                )
+                                .map_err(|e| Error::from_milli(e, Some(uuid.to_string())))?;
                         }
                         Available(index) => break index,
                         Closing(_) => {
@@ -460,6 +470,7 @@ impl IndexMapper {
                 let index = self.index(rtxn, index_uid)?;
                 let index_rtxn = index.read_txn()?;
                 IndexStats::new(&index, &index_rtxn)
+                    .map_err(|e| Error::from_milli(e, Some(uuid.to_string())))
             }
         }
     }

crates/index-scheduler/src/insta_snapshot.rs

@@ -353,7 +353,7 @@ pub fn snapshot_canceled_by(rtxn: &RoTxn, db: Database<BEU32, RoaringBitmapCodec
 
 pub fn snapshot_batch(batch: &Batch) -> String {
     let mut snap = String::new();
-    let Batch { uid, details, stats, started_at, finished_at } = batch;
+    let Batch { uid, details, stats, started_at, finished_at, progress: _ } = batch;
     if let Some(finished_at) = finished_at {
         assert!(finished_at > started_at);
     }

crates/index-scheduler/src/lib.rs

@@ -26,6 +26,7 @@ mod index_mapper;
 #[cfg(test)]
 mod insta_snapshot;
 mod lru;
+mod processing;
 mod utils;
 pub mod uuid_codec;
 
@@ -54,14 +55,13 @@ use meilisearch_types::features::{InstanceTogglableFeatures, RuntimeTogglableFea
 use meilisearch_types::heed::byteorder::BE;
 use meilisearch_types::heed::types::{SerdeBincode, SerdeJson, Str, I128};
 use meilisearch_types::heed::{self, Database, Env, PutFlags, RoTxn, RwTxn};
-use meilisearch_types::milli::documents::DocumentsBatchBuilder;
 use meilisearch_types::milli::index::IndexEmbeddingConfig;
-use meilisearch_types::milli::update::new::indexer::document_changes::Progress;
 use meilisearch_types::milli::update::IndexerConfig;
 use meilisearch_types::milli::vector::{Embedder, EmbedderOptions, EmbeddingConfigs};
 use meilisearch_types::milli::{self, CboRoaringBitmapCodec, Index, RoaringBitmapCodec, BEU32};
 use meilisearch_types::task_view::TaskView;
-use meilisearch_types::tasks::{Kind, KindWithContent, Status, Task, TaskProgress};
+use meilisearch_types::tasks::{Kind, KindWithContent, Status, Task};
+use processing::ProcessingTasks;
 use rayon::current_num_threads;
 use rayon::prelude::{IntoParallelIterator, ParallelIterator};
 use roaring::RoaringBitmap;
@@ -72,7 +72,8 @@ use utils::{filter_out_references_to_newer_tasks, keep_ids_within_datetimes, map
 use uuid::Uuid;
 
 use crate::index_mapper::IndexMapper;
-use crate::utils::{check_index_swap_validity, clamp_to_page_size, ProcessingBatch};
+use crate::processing::{AtomicTaskStep, BatchProgress};
+use crate::utils::{check_index_swap_validity, clamp_to_page_size};
 
 pub(crate) type BEI128 = I128<BE>;
 
@@ -163,48 +164,6 @@ impl Query {
     }
 }
 
-#[derive(Debug, Clone)]
-pub struct ProcessingTasks {
-    batch: Option<ProcessingBatch>,
-    /// The list of tasks ids that are currently running.
-    processing: RoaringBitmap,
-    /// The progress on processing tasks
-    progress: Option<TaskProgress>,
-}
-
-impl ProcessingTasks {
-    /// Creates an empty `ProcessingAt` struct.
-    fn new() -> ProcessingTasks {
-        ProcessingTasks { batch: None, processing: RoaringBitmap::new(), progress: None }
-    }
-
-    /// Stores the currently processing tasks, and the date time at which it started.
-    fn start_processing(&mut self, processing_batch: ProcessingBatch, processing: RoaringBitmap) {
-        self.batch = Some(processing_batch);
-        self.processing = processing;
-    }
-
-    fn update_progress(&mut self, progress: Progress) -> TaskProgress {
-        self.progress.get_or_insert_with(TaskProgress::default).update(progress)
-    }
-
-    /// Set the processing tasks to an empty list
-    fn stop_processing(&mut self) -> Self {
-        self.progress = None;
-
-        Self {
-            batch: std::mem::take(&mut self.batch),
-            processing: std::mem::take(&mut self.processing),
-            progress: None,
-        }
-    }
-
-    /// Returns `true` if there, at least, is one task that is currently processing that we must stop.
-    fn must_cancel_processing_tasks(&self, canceled_tasks: &RoaringBitmap) -> bool {
-        !self.processing.is_disjoint(canceled_tasks)
-    }
-}
-
 #[derive(Default, Clone, Debug)]
 struct MustStopProcessing(Arc<AtomicBool>);
 
@@ -407,7 +366,7 @@ pub struct IndexScheduler {
     ///
     /// See [self.breakpoint()](`IndexScheduler::breakpoint`) for an explanation.
     #[cfg(test)]
-    test_breakpoint_sdr: crossbeam::channel::Sender<(Breakpoint, bool)>,
+    test_breakpoint_sdr: crossbeam_channel::Sender<(Breakpoint, bool)>,
 
     /// A list of planned failures within the [`tick`](IndexScheduler::tick) method of the index scheduler.
     ///
@@ -476,7 +435,7 @@ impl IndexScheduler {
     /// Create an index scheduler and start its run loop.
     pub fn new(
         options: IndexSchedulerOptions,
-        #[cfg(test)] test_breakpoint_sdr: crossbeam::channel::Sender<(Breakpoint, bool)>,
+        #[cfg(test)] test_breakpoint_sdr: crossbeam_channel::Sender<(Breakpoint, bool)>,
         #[cfg(test)] planned_failures: Vec<(usize, tests::FailureLocation)>,
     ) -> Result<Self> {
         std::fs::create_dir_all(&options.tasks_path)?;
@@ -813,7 +772,7 @@ impl IndexScheduler {
             let mut batch_tasks = RoaringBitmap::new();
             for batch_uid in batch_uids {
                 if processing_batch.as_ref().map_or(false, |batch| batch.uid == *batch_uid) {
-                    batch_tasks |= &processing_tasks;
+                    batch_tasks |= &*processing_tasks;
                 } else {
                     batch_tasks |= self.tasks_in_batch(rtxn, *batch_uid)?;
                 }
@@ -827,13 +786,13 @@ impl IndexScheduler {
                 match status {
                     // special case for Processing tasks
                     Status::Processing => {
-                        status_tasks |= &processing_tasks;
+                        status_tasks |= &*processing_tasks;
                     }
                     status => status_tasks |= &self.get_status(rtxn, *status)?,
                 };
             }
             if !status.contains(&Status::Processing) {
-                tasks -= &processing_tasks;
+                tasks -= &*processing_tasks;
             }
             tasks &= status_tasks;
         }
@@ -882,7 +841,7 @@ impl IndexScheduler {
         // Once we have filtered the two subsets, we put them back together and assign it back to `tasks`.
         tasks = {
             let (mut filtered_non_processing_tasks, mut filtered_processing_tasks) =
-                (&tasks - &processing_tasks, &tasks & &processing_tasks);
+                (&tasks - &*processing_tasks, &tasks & &*processing_tasks);
 
             // special case for Processing tasks
             // A closure that clears the filtered_processing_tasks if their started_at date falls outside the given bounds
@@ -1090,7 +1049,7 @@ impl IndexScheduler {
         // Once we have filtered the two subsets, we put them back together and assign it back to `batches`.
         batches = {
             let (mut filtered_non_processing_batches, mut filtered_processing_batches) =
-                (&batches - &processing.processing, &batches & &processing.processing);
+                (&batches - &*processing.processing, &batches & &*processing.processing);
 
             // special case for Processing batches
             // A closure that clears the filtered_processing_batches if their started_at date falls outside the given bounds
@@ -1440,7 +1399,7 @@ impl IndexScheduler {
 
         // if the task doesn't delete anything and 50% of the task queue is full, we must refuse to enqueue the incomming task
         if !matches!(&kind, KindWithContent::TaskDeletion { tasks, .. } if !tasks.is_empty())
-            && (self.env.non_free_pages_size()? * 100) / self.env.info().map_size as u64 > 50
+            && (self.env.non_free_pages_size()? * 100) / self.env.info().map_size as u64 > 40
         {
             return Err(Error::NoSpaceLeftInTaskQueue);
         }
@@ -1606,7 +1565,8 @@ impl IndexScheduler {
 
         // We reset the must_stop flag to be sure that we don't stop processing tasks
         self.must_stop_processing.reset();
-        self.processing_tasks
+        let progress = self
+            .processing_tasks
             .write()
             .unwrap()
             // We can clone the processing batch here because we don't want its modification to affect the view of the processing batches
@@ -1619,11 +1579,12 @@ impl IndexScheduler {
         let res = {
             let cloned_index_scheduler = self.private_clone();
             let processing_batch = &mut processing_batch;
+            let progress = progress.clone();
             std::thread::scope(|s| {
                 let handle = std::thread::Builder::new()
                     .name(String::from("batch-operation"))
                     .spawn_scoped(s, move || {
-                        cloned_index_scheduler.process_batch(batch, processing_batch)
+                        cloned_index_scheduler.process_batch(batch, processing_batch, progress)
                     })
                     .unwrap();
                 handle.join().unwrap_or(Err(Error::ProcessBatchPanicked))
@@ -1636,6 +1597,7 @@ impl IndexScheduler {
         #[cfg(test)]
         self.maybe_fail(tests::FailureLocation::AcquiringWtxn)?;
 
+        progress.update_progress(BatchProgress::WritingTasksToDisk);
         processing_batch.finished();
         let mut wtxn = self.env.write_txn().map_err(Error::HeedTransaction)?;
         let mut canceled = RoaringBitmap::new();
@@ -1645,12 +1607,15 @@ impl IndexScheduler {
                 #[cfg(test)]
                 self.breakpoint(Breakpoint::ProcessBatchSucceeded);
 
+                let (task_progress, task_progress_obj) = AtomicTaskStep::new(tasks.len() as u32);
+                progress.update_progress(task_progress_obj);
                 let mut success = 0;
                 let mut failure = 0;
                 let mut canceled_by = None;
 
                 #[allow(unused_variables)]
                 for (i, mut task) in tasks.into_iter().enumerate() {
+                    task_progress.fetch_add(1, Ordering::Relaxed);
                     processing_batch.update(&mut task);
                     if task.status == Status::Canceled {
                         canceled.insert(task.uid);
@@ -1678,9 +1643,10 @@ impl IndexScheduler {
                 tracing::info!("A batch of tasks was successfully completed with {success} successful tasks and {failure} failed tasks.");
             }
             // If we have an abortion error we must stop the tick here and re-schedule tasks.
-            Err(Error::Milli(milli::Error::InternalError(
-                milli::InternalError::AbortedIndexation,
-            )))
+            Err(Error::Milli {
+                error: milli::Error::InternalError(milli::InternalError::AbortedIndexation),
+                ..
+            })
             | Err(Error::AbortedTask) => {
                 #[cfg(test)]
                 self.breakpoint(Breakpoint::AbortedIndexation);
@@ -1699,9 +1665,10 @@ impl IndexScheduler {
             // 2. close the associated environment
             // 3. resize it
             // 4. re-schedule tasks
-            Err(Error::Milli(milli::Error::UserError(
-                milli::UserError::MaxDatabaseSizeReached,
-            ))) if index_uid.is_some() => {
+            Err(Error::Milli {
+                error: milli::Error::UserError(milli::UserError::MaxDatabaseSizeReached),
+                ..
+            }) if index_uid.is_some() => {
                 // fixme: add index_uid to match to avoid the unwrap
                 let index_uid = index_uid.unwrap();
                 // fixme: handle error more gracefully? not sure when this could happen
@@ -1716,8 +1683,12 @@ impl IndexScheduler {
             Err(err) => {
                 #[cfg(test)]
                 self.breakpoint(Breakpoint::ProcessBatchFailed);
+                let (task_progress, task_progress_obj) = AtomicTaskStep::new(ids.len() as u32);
+                progress.update_progress(task_progress_obj);
+
                 let error: ResponseError = err.into();
                 for id in ids.iter() {
+                    task_progress.fetch_add(1, Ordering::Relaxed);
                     let mut task = self
                         .get_task(&wtxn, id)
                         .map_err(|e| Error::TaskDatabaseUpdate(Box::new(e)))?
@@ -1738,11 +1709,8 @@ impl IndexScheduler {
             }
         }
 
-        self.processing_tasks.write().unwrap().stop_processing();
         // We must re-add the canceled task so they're part of the same batch.
-        // processed.processing |= canceled;
         ids |= canceled;
-
         self.write_batch(&mut wtxn, processing_batch, &ids)?;
 
         #[cfg(test)]
@@ -1750,6 +1718,10 @@ impl IndexScheduler {
 
         wtxn.commit().map_err(Error::HeedTransaction)?;
 
+        // We should stop processing AFTER everything is processed and written to disk otherwise, a batch (which only lives in RAM) may appear in the processing task
+        // and then become « not found » for some time until the commit everything is written and the final commit is made.
+        self.processing_tasks.write().unwrap().stop_processing();
+
         // Once the tasks are committed, we should delete all the update files associated ASAP to avoid leaking files in case of a restart
         tracing::debug!("Deleting the update files");
 
@@ -1942,6 +1914,7 @@ impl IndexScheduler {
     // TODO: consider using a type alias or a struct embedder/template
     pub fn embedders(
         &self,
+        index_uid: String,
         embedding_configs: Vec<IndexEmbeddingConfig>,
     ) -> Result<EmbeddingConfigs> {
         let res: Result<_> = embedding_configs
@@ -1952,8 +1925,12 @@ impl IndexScheduler {
                      config: milli::vector::EmbeddingConfig { embedder_options, prompt, quantized },
                      ..
                  }| {
-                    let prompt =
-                        Arc::new(prompt.try_into().map_err(meilisearch_types::milli::Error::from)?);
+                    let prompt = Arc::new(
+                        prompt
+                            .try_into()
+                            .map_err(meilisearch_types::milli::Error::from)
+                            .map_err(|err| Error::from_milli(err, Some(index_uid.clone())))?,
+                    );
                     // optimistically return existing embedder
                     {
                         let embedders = self.embedders.read().unwrap();
@@ -1969,7 +1946,9 @@ impl IndexScheduler {
                     let embedder = Arc::new(
                         Embedder::new(embedder_options.clone())
                             .map_err(meilisearch_types::milli::vector::Error::from)
-                            .map_err(meilisearch_types::milli::Error::from)?,
+                            .map_err(|err| {
+                                Error::from_milli(err.into(), Some(index_uid.clone()))
+                            })?,
                     );
                     {
                         let mut embedders = self.embedders.write().unwrap();
@@ -2037,14 +2016,21 @@ impl<'a> Dump<'a> {
         task: TaskDump,
         content_file: Option<Box<UpdateFile>>,
     ) -> Result<Task> {
+        let task_has_no_docs = matches!(task.kind, KindDump::DocumentImport { documents_count, .. } if documents_count == 0);
+
         let content_uuid = match content_file {
             Some(content_file) if task.status == Status::Enqueued => {
-                let (uuid, mut file) = self.index_scheduler.create_update_file(false)?;
-                let mut builder = DocumentsBatchBuilder::new(&mut file);
+                let (uuid, file) = self.index_scheduler.create_update_file(false)?;
+                let mut writer = io::BufWriter::new(file);
                 for doc in content_file {
-                    builder.append_json_object(&doc?)?;
+                    let doc = doc?;
+                    serde_json::to_writer(&mut writer, &doc)
+                        .map_err(|e| {
+                            Error::from_milli(milli::InternalError::SerdeJson(e).into(), None)
+                        })
+                        .unwrap();
                 }
-                builder.into_inner()?;
+                let file = writer.into_inner().map_err(|e| e.into_error())?;
                 file.persist()?;
 
                 Some(uuid)
@@ -2052,6 +2038,12 @@ impl<'a> Dump<'a> {
             // If the task isn't `Enqueued` then just generate a recognisable `Uuid`
             // in case we try to open it later.
             _ if task.status != Status::Enqueued => Some(Uuid::nil()),
+            None if task.status == Status::Enqueued && task_has_no_docs => {
+                let (uuid, file) = self.index_scheduler.create_update_file(false)?;
+                file.persist()?;
+
+                Some(uuid)
+            }
             _ => None,
         };
 
@@ -2237,7 +2229,7 @@ mod tests {
     use std::time::Instant;
 
     use big_s::S;
-    use crossbeam::channel::RecvTimeoutError;
+    use crossbeam_channel::RecvTimeoutError;
     use file_store::File;
     use insta::assert_json_snapshot;
     use maplit::btreeset;
@@ -2289,7 +2281,7 @@ mod tests {
             configuration: impl Fn(&mut IndexSchedulerOptions),
         ) -> (Self, IndexSchedulerHandle) {
             let tempdir = TempDir::new().unwrap();
-            let (sender, receiver) = crossbeam::channel::bounded(0);
+            let (sender, receiver) = crossbeam_channel::bounded(0);
 
             let indexer_config = IndexerConfig { skip_index_budget: true, ..Default::default() };
 
@@ -2421,7 +2413,7 @@ mod tests {
     pub struct IndexSchedulerHandle {
         _tempdir: TempDir,
         index_scheduler: IndexScheduler,
-        test_breakpoint_rcv: crossbeam::channel::Receiver<(Breakpoint, bool)>,
+        test_breakpoint_rcv: crossbeam_channel::Receiver<(Breakpoint, bool)>,
         last_breakpoint: Breakpoint,
     }
 
@@ -4318,10 +4310,35 @@ mod tests {
         let proc = index_scheduler.processing_tasks.read().unwrap().clone();
 
         let query = Query { statuses: Some(vec![Status::Processing]), ..Default::default() };
-        let (batches, _) = index_scheduler
-            .get_batch_ids_from_authorized_indexes(&rtxn, &proc, &query, &AuthFilter::default())
+        let (mut batches, _) = index_scheduler
+            .get_batches_from_authorized_indexes(query.clone(), &AuthFilter::default())
             .unwrap();
-        snapshot!(snapshot_bitmap(&batches), @"[0,]"); // only the processing batch in the first tick
+        assert_eq!(batches.len(), 1);
+        batches[0].started_at = OffsetDateTime::UNIX_EPOCH;
+        // Insta cannot snapshot our batches because the batch stats contains an enum as key: https://github.com/mitsuhiko/insta/issues/689
+        let batch = serde_json::to_string_pretty(&batches[0]).unwrap();
+        snapshot!(batch, @r#"
+        {
+          "uid": 0,
+          "details": {
+            "primaryKey": "mouse"
+          },
+          "stats": {
+            "totalNbTasks": 1,
+            "status": {
+              "processing": 1
+            },
+            "types": {
+              "indexCreation": 1
+            },
+            "indexUids": {
+              "catto": 1
+            }
+          },
+          "startedAt": "1970-01-01T00:00:00Z",
+          "finishedAt": null
+        }
+        "#);
 
         let query = Query { statuses: Some(vec![Status::Enqueued]), ..Default::default() };
         let (batches, _) = index_scheduler
@@ -6145,7 +6162,7 @@ mod tests {
             insta::assert_json_snapshot!(simple_hf_config.embedder_options);
             let simple_hf_name = name.clone();
 
-            let configs = index_scheduler.embedders(configs).unwrap();
+            let configs = index_scheduler.embedders("doggos".to_string(), configs).unwrap();
             let (hf_embedder, _, _) = configs.get(&simple_hf_name).unwrap();
             let beagle_embed =
                 hf_embedder.embed_one(S("Intel the beagle best doggo"), None).unwrap();

crates/index-scheduler/src/processing.rs

@@ -0,0 +1,316 @@
+use std::borrow::Cow;
+use std::sync::Arc;
+
+use enum_iterator::Sequence;
+use meilisearch_types::milli::progress::{AtomicSubStep, NamedStep, Progress, ProgressView, Step};
+use meilisearch_types::milli::{make_atomic_progress, make_enum_progress};
+use roaring::RoaringBitmap;
+
+use crate::utils::ProcessingBatch;
+
+#[derive(Clone)]
+pub struct ProcessingTasks {
+    pub batch: Option<Arc<ProcessingBatch>>,
+    /// The list of tasks ids that are currently running.
+    pub processing: Arc<RoaringBitmap>,
+    /// The progress on processing tasks
+    pub progress: Option<Progress>,
+}
+
+impl ProcessingTasks {
+    /// Creates an empty `ProcessingAt` struct.
+    pub fn new() -> ProcessingTasks {
+        ProcessingTasks { batch: None, processing: Arc::new(RoaringBitmap::new()), progress: None }
+    }
+
+    pub fn get_progress_view(&self) -> Option<ProgressView> {
+        Some(self.progress.as_ref()?.as_progress_view())
+    }
+
+    /// Stores the currently processing tasks, and the date time at which it started.
+    pub fn start_processing(
+        &mut self,
+        processing_batch: ProcessingBatch,
+        processing: RoaringBitmap,
+    ) -> Progress {
+        self.batch = Some(Arc::new(processing_batch));
+        self.processing = Arc::new(processing);
+        let progress = Progress::default();
+        progress.update_progress(BatchProgress::ProcessingTasks);
+        self.progress = Some(progress.clone());
+
+        progress
+    }
+
+    /// Set the processing tasks to an empty list
+    pub fn stop_processing(&mut self) -> Self {
+        self.progress = None;
+
+        Self {
+            batch: std::mem::take(&mut self.batch),
+            processing: std::mem::take(&mut self.processing),
+            progress: None,
+        }
+    }
+
+    /// Returns `true` if there, at least, is one task that is currently processing that we must stop.
+    pub fn must_cancel_processing_tasks(&self, canceled_tasks: &RoaringBitmap) -> bool {
+        !self.processing.is_disjoint(canceled_tasks)
+    }
+}
+
+make_enum_progress! {
+    pub enum BatchProgress {
+        ProcessingTasks,
+        WritingTasksToDisk,
+    }
+}
+
+make_enum_progress! {
+    pub enum TaskCancelationProgress {
+        RetrievingTasks,
+        UpdatingTasks,
+    }
+}
+
+make_enum_progress! {
+    pub enum TaskDeletionProgress {
+        DeletingTasksDateTime,
+        DeletingTasksMetadata,
+        DeletingTasks,
+        DeletingBatches,
+    }
+}
+
+make_enum_progress! {
+    pub enum SnapshotCreationProgress {
+        StartTheSnapshotCreation,
+        SnapshotTheIndexScheduler,
+        SnapshotTheUpdateFiles,
+        SnapshotTheIndexes,
+        SnapshotTheApiKeys,
+        CreateTheTarball,
+    }
+}
+
+make_enum_progress! {
+    pub enum DumpCreationProgress {
+        StartTheDumpCreation,
+        DumpTheApiKeys,
+        DumpTheTasks,
+        DumpTheIndexes,
+        DumpTheExperimentalFeatures,
+        CompressTheDump,
+    }
+}
+
+make_enum_progress! {
+    pub enum CreateIndexProgress {
+        CreatingTheIndex,
+    }
+}
+
+make_enum_progress! {
+    pub enum UpdateIndexProgress {
+        UpdatingTheIndex,
+    }
+}
+
+make_enum_progress! {
+    pub enum DeleteIndexProgress {
+        DeletingTheIndex,
+    }
+}
+
+make_enum_progress! {
+    pub enum SwappingTheIndexes {
+        EnsuringCorrectnessOfTheSwap,
+        SwappingTheIndexes,
+    }
+}
+
+make_enum_progress! {
+    pub enum InnerSwappingTwoIndexes {
+        RetrieveTheTasks,
+        UpdateTheTasks,
+        UpdateTheIndexesMetadata,
+    }
+}
+
+make_enum_progress! {
+    pub enum DocumentOperationProgress {
+        RetrievingConfig,
+        ComputingDocumentChanges,
+        Indexing,
+    }
+}
+
+make_enum_progress! {
+    pub enum DocumentEditionProgress {
+        RetrievingConfig,
+        ComputingDocumentChanges,
+        Indexing,
+    }
+}
+
+make_enum_progress! {
+    pub enum DocumentDeletionProgress {
+        RetrievingConfig,
+        DeleteDocuments,
+        Indexing,
+    }
+}
+
+make_enum_progress! {
+    pub enum SettingsProgress {
+        RetrievingAndMergingTheSettings,
+        ApplyTheSettings,
+    }
+}
+
+make_atomic_progress!(Task alias AtomicTaskStep => "task" );
+make_atomic_progress!(Document alias AtomicDocumentStep => "document" );
+make_atomic_progress!(Batch alias AtomicBatchStep => "batch" );
+make_atomic_progress!(UpdateFile alias AtomicUpdateFileStep => "update file" );
+
+pub struct VariableNameStep {
+    name: String,
+    current: u32,
+    total: u32,
+}
+
+impl VariableNameStep {
+    pub fn new(name: impl Into<String>, current: u32, total: u32) -> Self {
+        Self { name: name.into(), current, total }
+    }
+}
+
+impl Step for VariableNameStep {
+    fn name(&self) -> Cow<'static, str> {
+        self.name.clone().into()
+    }
+
+    fn current(&self) -> u32 {
+        self.current
+    }
+
+    fn total(&self) -> u32 {
+        self.total
+    }
+}
+
+#[cfg(test)]
+mod test {
+    use std::sync::atomic::Ordering;
+
+    use meili_snap::{json_string, snapshot};
+
+    use super::*;
+
+    #[test]
+    fn one_level() {
+        let mut processing = ProcessingTasks::new();
+        processing.start_processing(ProcessingBatch::new(0), RoaringBitmap::new());
+        snapshot!(json_string!(processing.get_progress_view()), @r#"
+        {
+          "steps": [
+            {
+              "currentStep": "processing tasks",
+              "finished": 0,
+              "total": 2
+            }
+          ],
+          "percentage": 0.0
+        }
+        "#);
+        processing.progress.as_ref().unwrap().update_progress(BatchProgress::WritingTasksToDisk);
+        snapshot!(json_string!(processing.get_progress_view()), @r#"
+        {
+          "steps": [
+            {
+              "currentStep": "writing tasks to disk",
+              "finished": 1,
+              "total": 2
+            }
+          ],
+          "percentage": 50.0
+        }
+        "#);
+    }
+
+    #[test]
+    fn task_progress() {
+        let mut processing = ProcessingTasks::new();
+        processing.start_processing(ProcessingBatch::new(0), RoaringBitmap::new());
+        let (atomic, tasks) = AtomicTaskStep::new(10);
+        processing.progress.as_ref().unwrap().update_progress(tasks);
+        snapshot!(json_string!(processing.get_progress_view()), @r#"
+        {
+          "steps": [
+            {
+              "currentStep": "processing tasks",
+              "finished": 0,
+              "total": 2
+            },
+            {
+              "currentStep": "task",
+              "finished": 0,
+              "total": 10
+            }
+          ],
+          "percentage": 0.0
+        }
+        "#);
+        atomic.fetch_add(6, Ordering::Relaxed);
+        snapshot!(json_string!(processing.get_progress_view()), @r#"
+        {
+          "steps": [
+            {
+              "currentStep": "processing tasks",
+              "finished": 0,
+              "total": 2
+            },
+            {
+              "currentStep": "task",
+              "finished": 6,
+              "total": 10
+            }
+          ],
+          "percentage": 30.000002
+        }
+        "#);
+        processing.progress.as_ref().unwrap().update_progress(BatchProgress::WritingTasksToDisk);
+        snapshot!(json_string!(processing.get_progress_view()), @r#"
+        {
+          "steps": [
+            {
+              "currentStep": "writing tasks to disk",
+              "finished": 1,
+              "total": 2
+            }
+          ],
+          "percentage": 50.0
+        }
+        "#);
+        let (atomic, tasks) = AtomicTaskStep::new(5);
+        processing.progress.as_ref().unwrap().update_progress(tasks);
+        atomic.fetch_add(4, Ordering::Relaxed);
+        snapshot!(json_string!(processing.get_progress_view()), @r#"
+        {
+          "steps": [
+            {
+              "currentStep": "writing tasks to disk",
+              "finished": 1,
+              "total": 2
+            },
+            {
+              "currentStep": "task",
+              "finished": 4,
+              "total": 5
+            }
+          ],
+          "percentage": 90.0
+        }
+        "#);
+    }
+}

crates/index-scheduler/src/snapshots/lib.rs/cancel_mix_of_tasks/aborted_indexation.snap

@@ -5,7 +5,7 @@ snapshot_kind: text
 ### Autobatching Enabled = true
 ### Processing batch Some(1):
 [1,]
-{uid: 1, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"beavero":2}}, }
+{uid: 1, details: {"receivedDocuments":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"beavero":1}}, }
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, batch_uid: 0, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "catto", primary_key: None, method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}

crates/index-scheduler/src/snapshots/lib.rs/cancel_mix_of_tasks/processing_second_task_cancel_enqueued.snap

@@ -5,7 +5,7 @@ snapshot_kind: text
 ### Autobatching Enabled = true
 ### Processing batch Some(1):
 [1,]
-{uid: 1, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"beavero":2}}, }
+{uid: 1, details: {"receivedDocuments":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"beavero":1}}, }
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, batch_uid: 0, status: succeeded, details: { received_documents: 1, indexed_documents: Some(1) }, kind: DocumentAdditionOrUpdate { index_uid: "catto", primary_key: None, method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}

crates/index-scheduler/src/snapshots/lib.rs/cancel_processing_dump/cancel_registered.snap

@@ -5,7 +5,7 @@ snapshot_kind: text
 ### Autobatching Enabled = true
 ### Processing batch Some(0):
 [0,]
-{uid: 0, details: {"dumpUid":null}, stats: {"totalNbTasks":1,"status":{"enqueued":1},"types":{"dumpCreation":1},"indexUids":{}}, }
+{uid: 0, details: {"dumpUid":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"dumpCreation":1},"indexUids":{}}, }
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, status: enqueued, details: { dump_uid: None }, kind: DumpCreation { keys: [], instance_uid: None }}

crates/index-scheduler/src/snapshots/lib.rs/cancel_processing_task/aborted_indexation.snap

@@ -5,7 +5,7 @@ snapshot_kind: text
 ### Autobatching Enabled = true
 ### Processing batch Some(0):
 [0,]
-{uid: 0, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"catto":2}}, }
+{uid: 0, details: {"receivedDocuments":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"catto":1}}, }
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, status: enqueued, details: { received_documents: 1, indexed_documents: None }, kind: DocumentAdditionOrUpdate { index_uid: "catto", primary_key: None, method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}

crates/index-scheduler/src/snapshots/lib.rs/cancel_processing_task/cancel_task_registered.snap

@@ -5,7 +5,7 @@ snapshot_kind: text
 ### Autobatching Enabled = true
 ### Processing batch Some(0):
 [0,]
-{uid: 0, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"catto":2}}, }
+{uid: 0, details: {"receivedDocuments":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"catto":1}}, }
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, status: enqueued, details: { received_documents: 1, indexed_documents: None }, kind: DocumentAdditionOrUpdate { index_uid: "catto", primary_key: None, method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}

crates/index-scheduler/src/snapshots/lib.rs/cancel_processing_task/initial_task_processing.snap

@@ -5,7 +5,7 @@ snapshot_kind: text
 ### Autobatching Enabled = true
 ### Processing batch Some(0):
 [0,]
-{uid: 0, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"catto":2}}, }
+{uid: 0, details: {"receivedDocuments":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"catto":1}}, }
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, status: enqueued, details: { received_documents: 1, indexed_documents: None }, kind: DocumentAdditionOrUpdate { index_uid: "catto", primary_key: None, method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}

crates/index-scheduler/src/snapshots/lib.rs/document_addition/after_the_batch_creation.snap

@@ -5,7 +5,7 @@ snapshot_kind: text
 ### Autobatching Enabled = true
 ### Processing batch Some(0):
 [0,]
-{uid: 0, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"doggos":2}}, }
+{uid: 0, details: {"receivedDocuments":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, status: enqueued, details: { received_documents: 1, indexed_documents: None }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}

crates/index-scheduler/src/snapshots/lib.rs/fail_in_process_batch_for_document_addition/document_addition_batch_created.snap

@@ -5,7 +5,7 @@ snapshot_kind: text
 ### Autobatching Enabled = true
 ### Processing batch Some(0):
 [0,]
-{uid: 0, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"doggos":2}}, }
+{uid: 0, details: {"receivedDocuments":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, status: enqueued, details: { received_documents: 1, indexed_documents: None }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}

crates/index-scheduler/src/snapshots/lib.rs/fail_in_process_batch_for_document_deletion/after_removing_the_documents.snap

@@ -9,8 +9,8 @@ source: crates/index-scheduler/src/lib.rs
 0 {uid: 0, batch_uid: 0, status: succeeded, details: { settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, facet_search: NotSet, prefix_search: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> } }, kind: SettingsUpdate { index_uid: "doggos", new_settings: Settings { displayed_attributes: WildcardSetting(NotSet), searchable_attributes: WildcardSetting(NotSet), filterable_attributes: Set({"catto"}), sortable_attributes: NotSet, ranking_rules: NotSet, stop_words: NotSet, non_separator_tokens: NotSet, separator_tokens: NotSet, dictionary: NotSet, synonyms: NotSet, distinct_attribute: NotSet, proximity_precision: NotSet, typo_tolerance: NotSet, faceting: NotSet, pagination: NotSet, embedders: NotSet, search_cutoff_ms: NotSet, localized_attributes: NotSet, facet_search: NotSet, prefix_search: NotSet, _kind: PhantomData<meilisearch_types::settings::Unchecked> }, is_deletion: false, allow_index_creation: true }}
 1 {uid: 1, batch_uid: 1, status: succeeded, details: { received_documents: 3, indexed_documents: Some(3) }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 3, allow_index_creation: true }}
 2 {uid: 2, batch_uid: 2, status: succeeded, details: { received_document_ids: 1, deleted_documents: Some(1) }, kind: DocumentDeletion { index_uid: "doggos", documents_ids: ["1"] }}
-3 {uid: 3, batch_uid: 2, status: failed, error: ResponseError { code: 200, message: "Invalid type for filter subexpression: expected: String, Array, found: true.", error_code: "invalid_document_filter", error_type: "invalid_request", error_link: "https://docs.meilisearch.com/errors#invalid_document_filter" }, details: { original_filter: true, deleted_documents: Some(0) }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: Bool(true) }}
-4 {uid: 4, batch_uid: 2, status: failed, error: ResponseError { code: 200, message: "Attribute `id` is not filterable. Available filterable attributes are: `catto`.\n1:3 id = 2", error_code: "invalid_document_filter", error_type: "invalid_request", error_link: "https://docs.meilisearch.com/errors#invalid_document_filter" }, details: { original_filter: "id = 2", deleted_documents: Some(0) }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: String("id = 2") }}
+3 {uid: 3, batch_uid: 2, status: failed, error: ResponseError { code: 200, message: "Index `doggos`: Invalid type for filter subexpression: expected: String, Array, found: true.", error_code: "invalid_document_filter", error_type: "invalid_request", error_link: "https://docs.meilisearch.com/errors#invalid_document_filter" }, details: { original_filter: true, deleted_documents: Some(0) }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: Bool(true) }}
+4 {uid: 4, batch_uid: 2, status: failed, error: ResponseError { code: 200, message: "Index `doggos`: Attribute `id` is not filterable. Available filterable attributes are: `catto`.\n1:3 id = 2", error_code: "invalid_document_filter", error_type: "invalid_request", error_link: "https://docs.meilisearch.com/errors#invalid_document_filter" }, details: { original_filter: "id = 2", deleted_documents: Some(0) }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: String("id = 2") }}
 5 {uid: 5, batch_uid: 2, status: succeeded, details: { original_filter: "catto EXISTS", deleted_documents: Some(1) }, kind: DocumentDeletionByFilter { index_uid: "doggos", filter_expr: String("catto EXISTS") }}
 ----------------------------------------------------------------------
 ### Status:

crates/index-scheduler/src/snapshots/lib.rs/fail_in_update_task_after_process_batch_success_for_document_addition/after_batch_succeeded.snap

@@ -5,7 +5,7 @@ snapshot_kind: text
 ### Autobatching Enabled = true
 ### Processing batch Some(0):
 [0,]
-{uid: 0, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"doggos":2}}, }
+{uid: 0, details: {"receivedDocuments":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, status: enqueued, details: { received_documents: 1, indexed_documents: None }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}

crates/index-scheduler/src/snapshots/lib.rs/fail_in_update_task_after_process_batch_success_for_document_addition/after_failing_to_commit.snap

@@ -5,7 +5,7 @@ snapshot_kind: text
 ### Autobatching Enabled = true
 ### Processing batch Some(0):
 [0,]
-{uid: 0, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"doggos":2}}, }
+{uid: 0, details: {"receivedDocuments":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, status: enqueued, details: { received_documents: 1, indexed_documents: None }, kind: DocumentAdditionOrUpdate { index_uid: "doggos", primary_key: Some("id"), method: ReplaceDocuments, content_file: 00000000-0000-0000-0000-000000000000, documents_count: 1, allow_index_creation: true }}

crates/index-scheduler/src/snapshots/lib.rs/insert_task_while_another_task_is_processing/after_batch_creation.snap

@@ -5,7 +5,7 @@ snapshot_kind: text
 ### Autobatching Enabled = true
 ### Processing batch Some(0):
 [0,]
-{uid: 0, details: {"primaryKey":"id"}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"indexCreation":2},"indexUids":{"index_a":2}}, }
+{uid: 0, details: {"primaryKey":"id"}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"indexCreation":1},"indexUids":{"index_a":1}}, }
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, status: enqueued, details: { primary_key: Some("id") }, kind: IndexCreation { index_uid: "index_a", primary_key: Some("id") }}

crates/index-scheduler/src/snapshots/lib.rs/insert_task_while_another_task_is_processing/registered_the_second_task.snap

@@ -5,7 +5,7 @@ snapshot_kind: text
 ### Autobatching Enabled = true
 ### Processing batch Some(0):
 [0,]
-{uid: 0, details: {"primaryKey":"id"}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"indexCreation":2},"indexUids":{"index_a":2}}, }
+{uid: 0, details: {"primaryKey":"id"}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"indexCreation":1},"indexUids":{"index_a":1}}, }
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, status: enqueued, details: { primary_key: Some("id") }, kind: IndexCreation { index_uid: "index_a", primary_key: Some("id") }}

crates/index-scheduler/src/snapshots/lib.rs/insert_task_while_another_task_is_processing/registered_the_third_task.snap

@@ -5,7 +5,7 @@ snapshot_kind: text
 ### Autobatching Enabled = true
 ### Processing batch Some(0):
 [0,]
-{uid: 0, details: {"primaryKey":"id"}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"indexCreation":2},"indexUids":{"index_a":2}}, }
+{uid: 0, details: {"primaryKey":"id"}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"indexCreation":1},"indexUids":{"index_a":1}}, }
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, status: enqueued, details: { primary_key: Some("id") }, kind: IndexCreation { index_uid: "index_a", primary_key: Some("id") }}

crates/index-scheduler/src/snapshots/lib.rs/query_batches_simple/after-advancing-a-bit.snap

@@ -5,7 +5,7 @@ snapshot_kind: text
 ### Autobatching Enabled = true
 ### Processing batch Some(1):
 [1,]
-{uid: 1, details: {"primaryKey":"sheep"}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"indexCreation":2},"indexUids":{"doggo":2}}, }
+{uid: 1, details: {"primaryKey":"sheep"}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"indexCreation":1},"indexUids":{"doggo":1}}, }
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, batch_uid: 0, status: succeeded, details: { primary_key: Some("mouse") }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}

crates/index-scheduler/src/snapshots/lib.rs/task_deletion_undeleteable/task_deletion_processing.snap

@@ -5,7 +5,7 @@ snapshot_kind: text
 ### Autobatching Enabled = true
 ### Processing batch Some(0):
 [3,]
-{uid: 0, details: {"matchedTasks":2,"deletedTasks":null,"originalFilter":"test_query"}, stats: {"totalNbTasks":1,"status":{"enqueued":1},"types":{"taskDeletion":1},"indexUids":{}}, }
+{uid: 0, details: {"matchedTasks":2,"deletedTasks":null,"originalFilter":"test_query"}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"taskDeletion":1},"indexUids":{}}, }
 ----------------------------------------------------------------------
 ### All Tasks:
 0 {uid: 0, status: enqueued, details: { primary_key: Some("mouse") }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}

crates/index-scheduler/src/utils.rs

@@ -67,7 +67,7 @@ impl ProcessingBatch {
             task.batch_uid = Some(self.uid);
             // We don't store the statuses in the map since they're all enqueued but we must
             // still store them in the stats since that can be displayed.
-            *self.stats.status.entry(task.status).or_default() += 1;
+            *self.stats.status.entry(Status::Processing).or_default() += 1;
 
             self.kinds.insert(task.kind.as_kind());
             *self.stats.types.entry(task.kind.as_kind()).or_default() += 1;
@@ -106,7 +106,7 @@ impl ProcessingBatch {
         self.stats.total_nb_tasks = 0;
     }
 
-    /// Update the timestamp of the tasks and the inner structure of this sturcture.
+    /// Update the timestamp of the tasks and the inner structure of this structure.
     pub fn update(&mut self, task: &mut Task) {
         // We must re-set this value in case we're dealing with a task that has been added between
         // the `processing` and `finished` state
@@ -134,6 +134,7 @@ impl ProcessingBatch {
     pub fn to_batch(&self) -> Batch {
         Batch {
             uid: self.uid,
+            progress: None,
             details: self.details.clone(),
             stats: self.stats.clone(),
             started_at: self.started_at,
@@ -187,6 +188,7 @@ impl IndexScheduler {
             &batch.uid,
             &Batch {
                 uid: batch.uid,
+                progress: None,
                 details: batch.details,
                 stats: batch.stats,
                 started_at: batch.started_at,
@@ -273,7 +275,9 @@ impl IndexScheduler {
             .into_iter()
             .map(|batch_id| {
                 if Some(batch_id) == processing.batch.as_ref().map(|batch| batch.uid) {
-                    Ok(processing.batch.as_ref().unwrap().to_batch())
+                    let mut batch = processing.batch.as_ref().unwrap().to_batch();
+                    batch.progress = processing.get_progress_view();
+                    Ok(batch)
                 } else {
                     self.get_batch(rtxn, batch_id)
                         .and_then(|task| task.ok_or(Error::CorruptedTaskQueue))
@@ -287,7 +291,10 @@ impl IndexScheduler {
 
         debug_assert!(old_task != *task);
         debug_assert_eq!(old_task.uid, task.uid);
-        debug_assert!(old_task.batch_uid.is_none() && task.batch_uid.is_some());
+        debug_assert!(
+            old_task.batch_uid.is_none() && task.batch_uid.is_some(),
+            "\n==> old: {old_task:?}\n==> new: {task:?}"
+        );
 
         if old_task.status != task.status {
             self.update_status(wtxn, old_task.status, |bitmap| {

crates/meilisearch-auth/Cargo.toml

@@ -17,7 +17,7 @@ hmac = "0.12.1"
 maplit = "1.0.2"
 meilisearch-types = { path = "../meilisearch-types" }
 rand = "0.8.5"
-roaring = { version = "0.10.6", features = ["serde"] }
+roaring = { version = "0.10.7", features = ["serde"] }
 serde = { version = "1.0.204", features = ["derive"] }
 serde_json = { version = "1.0.120", features = ["preserve_order"] }
 sha2 = "0.10.8"

crates/meilisearch-types/Cargo.toml

@@ -24,8 +24,9 @@ flate2 = "1.0.30"
 fst = "0.4.7"
 memmap2 = "0.9.4"
 milli = { path = "../milli" }
-raw-collections = { git = "https://github.com/meilisearch/raw-collections.git", version = "0.1.0" }
-roaring = { version = "0.10.6", features = ["serde"] }
+bumparaw-collections = "0.1.2"
+roaring = { version = "0.10.7", features = ["serde"] }
+rustc-hash = "2.1.0"
 serde = { version = "1.0.204", features = ["derive"] }
 serde-cs = "0.2.4"
 serde_json = "1.0.120"

crates/meilisearch-types/src/batch_view.rs

@@ -1,16 +1,16 @@
+use milli::progress::ProgressView;
 use serde::Serialize;
 use time::{Duration, OffsetDateTime};
 
-use crate::{
-    batches::{Batch, BatchId, BatchStats},
-    task_view::DetailsView,
-    tasks::serialize_duration,
-};
+use crate::batches::{Batch, BatchId, BatchStats};
+use crate::task_view::DetailsView;
+use crate::tasks::serialize_duration;
 
 #[derive(Debug, Clone, Serialize)]
 #[serde(rename_all = "camelCase")]
 pub struct BatchView {
     pub uid: BatchId,
+    pub progress: Option<ProgressView>,
     pub details: DetailsView,
     pub stats: BatchStats,
     #[serde(serialize_with = "serialize_duration", default)]
@@ -25,6 +25,7 @@ impl BatchView {
     pub fn from_batch(batch: &Batch) -> Self {
         Self {
             uid: batch.uid,
+            progress: batch.progress.clone(),
             details: batch.details.clone(),
             stats: batch.stats.clone(),
             duration: batch.finished_at.map(|finished_at| finished_at - batch.started_at),

crates/meilisearch-types/src/batches.rs

@@ -1,12 +1,11 @@
 use std::collections::BTreeMap;
 
+use milli::progress::ProgressView;
 use serde::{Deserialize, Serialize};
 use time::OffsetDateTime;
 
-use crate::{
-    task_view::DetailsView,
-    tasks::{Kind, Status},
-};
+use crate::task_view::DetailsView;
+use crate::tasks::{Kind, Status};
 
 pub type BatchId = u32;
 
@@ -15,6 +14,8 @@ pub type BatchId = u32;
 pub struct Batch {
     pub uid: BatchId,
 
+    #[serde(skip)]
+    pub progress: Option<ProgressView>,
     pub details: DetailsView,
     pub stats: BatchStats,
 

crates/meilisearch-types/src/document_formats.rs

@@ -4,10 +4,11 @@ use std::io::{self, BufWriter};
 use std::marker::PhantomData;
 
 use bumpalo::Bump;
+use bumparaw_collections::RawMap;
 use memmap2::Mmap;
 use milli::documents::Error;
 use milli::Object;
-use raw_collections::RawMap;
+use rustc_hash::FxBuildHasher;
 use serde::de::{SeqAccess, Visitor};
 use serde::{Deserialize, Deserializer};
 use serde_json::error::Category;
@@ -214,13 +215,13 @@ pub fn read_json(input: &File, output: impl io::Write) -> Result<u64> {
     // We memory map to be able to deserialize into a RawMap that
     // does not allocate when possible and only materialize the first/top level.
     let input = unsafe { Mmap::map(input).map_err(DocumentFormatError::Io)? };
-    let mut doc_alloc = Bump::with_capacity(1024 * 1024 * 1024); // 1MiB
+    let mut doc_alloc = Bump::with_capacity(1024 * 1024); // 1MiB
 
     let mut out = BufWriter::new(output);
     let mut deserializer = serde_json::Deserializer::from_slice(&input);
     let res = array_each(&mut deserializer, |obj: &RawValue| {
         doc_alloc.reset();
-        let map = RawMap::from_raw_value(obj, &doc_alloc)?;
+        let map = RawMap::from_raw_value_and_hasher(obj, FxBuildHasher, &doc_alloc)?;
         to_writer(&mut out, &map)
     });
     let count = match res {
@@ -250,26 +251,25 @@ pub fn read_json(input: &File, output: impl io::Write) -> Result<u64> {
     }
 }
 
-/// Reads NDJSON from file and write it in NDJSON in a file checking it along the way.
-pub fn read_ndjson(input: &File, output: impl io::Write) -> Result<u64> {
+/// Reads NDJSON from file and checks it.
+pub fn read_ndjson(input: &File) -> Result<u64> {
     // We memory map to be able to deserialize into a RawMap that
     // does not allocate when possible and only materialize the first/top level.
     let input = unsafe { Mmap::map(input).map_err(DocumentFormatError::Io)? };
-    let mut output = BufWriter::new(output);
-
     let mut bump = Bump::with_capacity(1024 * 1024);
 
     let mut count = 0;
     for result in serde_json::Deserializer::from_slice(&input).into_iter() {
         bump.reset();
-        count += 1;
-        result
-            .and_then(|raw: &RawValue| {
+        match result {
+            Ok(raw) => {
                 // try to deserialize as a map
-                let map = RawMap::from_raw_value(raw, &bump)?;
-                to_writer(&mut output, &map)
-            })
-            .map_err(|e| DocumentFormatError::from((PayloadType::Ndjson, e)))?;
+                RawMap::from_raw_value_and_hasher(raw, FxBuildHasher, &bump)
+                    .map_err(|e| DocumentFormatError::from((PayloadType::Ndjson, e)))?;
+                count += 1;
+            }
+            Err(e) => return Err(DocumentFormatError::from((PayloadType::Ndjson, e))),
+        }
     }
 
     Ok(count)

crates/meilisearch-types/src/error.rs

@@ -279,6 +279,7 @@ InvalidSearchPage                     , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchQ                        , InvalidRequest       , BAD_REQUEST ;
 InvalidFacetSearchQuery               , InvalidRequest       , BAD_REQUEST ;
 InvalidFacetSearchName                , InvalidRequest       , BAD_REQUEST ;
+FacetSearchDisabled                   , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchVector                   , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchShowMatchesPosition      , InvalidRequest       , BAD_REQUEST ;
 InvalidSearchShowRankingScore         , InvalidRequest       , BAD_REQUEST ;
@@ -549,7 +550,7 @@ impl fmt::Display for deserr_codes::InvalidSimilarId {
             "the value of `id` is invalid. \
             A document identifier can be of type integer or string, \
             only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_), \
-            and can not be more than 512 bytes."
+            and can not be more than 511 bytes."
         )
     }
 }

crates/meilisearch-types/src/tasks.rs

@@ -4,7 +4,6 @@ use std::fmt::{Display, Write};
 use std::str::FromStr;
 
 use enum_iterator::Sequence;
-use milli::update::new::indexer::document_changes::Progress;
 use milli::update::IndexDocumentsMethod;
 use milli::Object;
 use roaring::RoaringBitmap;
@@ -41,62 +40,6 @@ pub struct Task {
     pub kind: KindWithContent,
 }
 
-#[derive(Clone, Copy, Debug, PartialEq, Eq, Serialize, Deserialize)]
-#[serde(rename_all = "camelCase")]
-pub struct TaskProgress {
-    pub current_step: &'static str,
-    pub finished_steps: u16,
-    pub total_steps: u16,
-    pub finished_substeps: Option<u32>,
-    pub total_substeps: Option<u32>,
-}
-
-impl Default for TaskProgress {
-    fn default() -> Self {
-        Self::new()
-    }
-}
-
-impl TaskProgress {
-    pub fn new() -> Self {
-        Self {
-            current_step: "start",
-            finished_steps: 0,
-            total_steps: 1,
-            finished_substeps: None,
-            total_substeps: None,
-        }
-    }
-
-    pub fn update(&mut self, progress: Progress) -> TaskProgress {
-        if self.finished_steps > progress.finished_steps {
-            return *self;
-        }
-
-        if self.current_step != progress.step_name {
-            self.current_step = progress.step_name
-        }
-
-        self.total_steps = progress.total_steps;
-
-        if self.finished_steps < progress.finished_steps {
-            self.finished_substeps = None;
-            self.total_substeps = None;
-        }
-        self.finished_steps = progress.finished_steps;
-        if let Some((finished_substeps, total_substeps)) = progress.finished_total_substep {
-            if let Some(task_finished_substeps) = self.finished_substeps {
-                if task_finished_substeps > finished_substeps {
-                    return *self;
-                }
-            }
-            self.finished_substeps = Some(finished_substeps);
-            self.total_substeps = Some(total_substeps);
-        }
-        *self
-    }
-}
-
 impl Task {
     pub fn index_uid(&self) -> Option<&str> {
         use KindWithContent::*;

crates/meilisearch/Cargo.toml

@@ -103,7 +103,7 @@ tracing-subscriber = { version = "0.3.18", features = ["json"] }
 tracing-trace = { version = "0.1.0", path = "../tracing-trace" }
 tracing-actix-web = "0.7.11"
 build-info = { version = "1.7.0", path = "../build-info" }
-roaring = "0.10.2"
+roaring = "0.10.7"
 mopa-maintained = "0.2.3"
 
 [dev-dependencies]
@@ -157,5 +157,5 @@ german = ["meilisearch-types/german"]
 turkish = ["meilisearch-types/turkish"]
 
 [package.metadata.mini-dashboard]
-assets-url = "https://github.com/meilisearch/mini-dashboard/releases/download/v0.2.15/build.zip"
-sha1 = "d057600b4a839a2e0c0be7a372cd1b2683f3ca7e"
+assets-url = "https://github.com/meilisearch/mini-dashboard/releases/download/v0.2.16/build.zip"
+sha1 = "68f83438a114aabbe76bc9fe480071e741996662"

crates/meilisearch/src/error.rs

@@ -4,6 +4,7 @@ use byte_unit::{Byte, UnitType};
 use meilisearch_types::document_formats::{DocumentFormatError, PayloadType};
 use meilisearch_types::error::{Code, ErrorCode, ResponseError};
 use meilisearch_types::index_uid::{IndexUid, IndexUidFormatError};
+use meilisearch_types::milli;
 use meilisearch_types::milli::OrderBy;
 use serde_json::Value;
 use tokio::task::JoinError;
@@ -62,8 +63,11 @@ pub enum MeilisearchHttpError {
     HeedError(#[from] meilisearch_types::heed::Error),
     #[error(transparent)]
     IndexScheduler(#[from] index_scheduler::Error),
-    #[error(transparent)]
-    Milli(#[from] meilisearch_types::milli::Error),
+    #[error("{}", match .index_name {
+        Some(name) if !name.is_empty() => format!("Index `{}`: {error}", name),
+        _ => format!("{error}")
+    })]
+    Milli { error: milli::Error, index_name: Option<String> },
     #[error(transparent)]
     Payload(#[from] PayloadError),
     #[error(transparent)]
@@ -76,6 +80,12 @@ pub enum MeilisearchHttpError {
     MissingSearchHybrid,
 }
 
+impl MeilisearchHttpError {
+    pub(crate) fn from_milli(error: milli::Error, index_name: Option<String>) -> Self {
+        Self::Milli { error, index_name }
+    }
+}
+
 impl ErrorCode for MeilisearchHttpError {
     fn error_code(&self) -> Code {
         match self {
@@ -95,7 +105,7 @@ impl ErrorCode for MeilisearchHttpError {
             MeilisearchHttpError::SerdeJson(_) => Code::Internal,
             MeilisearchHttpError::HeedError(_) => Code::Internal,
             MeilisearchHttpError::IndexScheduler(e) => e.error_code(),
-            MeilisearchHttpError::Milli(e) => e.error_code(),
+            MeilisearchHttpError::Milli { error, .. } => error.error_code(),
             MeilisearchHttpError::Payload(e) => e.error_code(),
             MeilisearchHttpError::FileStore(_) => Code::Internal,
             MeilisearchHttpError::DocumentFormat(e) => e.error_code(),

crates/meilisearch/src/lib.rs

@@ -395,6 +395,7 @@ fn import_dump(
     for index_reader in dump_reader.indexes()? {
         let mut index_reader = index_reader?;
         let metadata = index_reader.metadata();
+        let uid = metadata.uid.clone();
         tracing::info!("Importing index `{}`.", metadata.uid);
 
         let date = Some((metadata.created_at, metadata.updated_at));
@@ -432,7 +433,7 @@ fn import_dump(
         let reader = DocumentsBatchReader::from_reader(reader)?;
 
         let embedder_configs = index.embedding_configs(&wtxn)?;
-        let embedders = index_scheduler.embedders(embedder_configs)?;
+        let embedders = index_scheduler.embedders(uid, embedder_configs)?;
 
         let builder = milli::update::IndexDocuments::new(
             &mut wtxn,

crates/meilisearch/src/main.rs

@@ -20,14 +20,14 @@ use meilisearch::{
     LogStderrType, Opt, SubscriberForSecondLayer,
 };
 use meilisearch_auth::{generate_master_key, AuthController, MASTER_KEY_MIN_SIZE};
-use mimalloc::MiMalloc;
 use termcolor::{Color, ColorChoice, ColorSpec, StandardStream, WriteColor};
 use tracing::level_filters::LevelFilter;
 use tracing_subscriber::layer::SubscriberExt as _;
 use tracing_subscriber::Layer;
 
+#[cfg(not(windows))]
 #[global_allocator]
-static ALLOC: MiMalloc = MiMalloc;
+static ALLOC: mimalloc::MiMalloc = mimalloc::MiMalloc;
 
 fn default_log_route_layer() -> LogRouteType {
     None.with_filter(tracing_subscriber::filter::Targets::new().with_target("", LevelFilter::OFF))
@@ -129,6 +129,11 @@ async fn try_main() -> anyhow::Result<()> {
 
     print_launch_resume(&opt, analytics.clone(), config_read_from);
 
+    tokio::spawn(async move {
+        tokio::signal::ctrl_c().await.unwrap();
+        std::process::exit(130);
+    });
+
     run_http(index_scheduler, auth_controller, opt, log_handle, Arc::new(analytics)).await?;
 
     Ok(())

crates/meilisearch/src/routes/batches.rs

@@ -1,18 +1,18 @@
-use actix_web::{
-    web::{self, Data},
-    HttpResponse,
-};
+use actix_web::web::{self, Data};
+use actix_web::HttpResponse;
 use deserr::actix_web::AwebQueryParameter;
 use index_scheduler::{IndexScheduler, Query};
-use meilisearch_types::{
-    batch_view::BatchView, batches::BatchId, deserr::DeserrQueryParamError, error::ResponseError,
-    keys::actions,
-};
+use meilisearch_types::batch_view::BatchView;
+use meilisearch_types::batches::BatchId;
+use meilisearch_types::deserr::DeserrQueryParamError;
+use meilisearch_types::error::ResponseError;
+use meilisearch_types::keys::actions;
 use serde::Serialize;
 
-use crate::extractors::{authentication::GuardedData, sequential_extractor::SeqHandler};
-
-use super::{tasks::TasksFilterQuery, ActionPolicy};
+use super::tasks::TasksFilterQuery;
+use super::ActionPolicy;
+use crate::extractors::authentication::GuardedData;
+use crate::extractors::sequential_extractor::SeqHandler;
 
 pub fn configure(cfg: &mut web::ServiceConfig) {
     cfg.service(web::resource("").route(web::get().to(SeqHandler(get_batches))))

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

@@ -1,5 +1,5 @@
 use std::collections::HashSet;
-use std::io::ErrorKind;
+use std::io::{ErrorKind, Seek as _};
 use std::marker::PhantomData;
 
 use actix_web::http::header::CONTENT_TYPE;
@@ -572,7 +572,7 @@ async fn document_addition(
     index_uid: IndexUid,
     primary_key: Option<String>,
     csv_delimiter: Option<u8>,
-    mut body: Payload,
+    body: Payload,
     method: IndexDocumentsMethod,
     task_id: Option<TaskId>,
     dry_run: bool,
@@ -609,54 +609,60 @@ async fn document_addition(
     };
 
     let (uuid, mut update_file) = index_scheduler.create_update_file(dry_run)?;
+    let documents_count = match format {
+        PayloadType::Ndjson => {
+            let (path, file) = update_file.into_parts();
+            let file = match file {
+                Some(file) => {
+                    let (file, path) = file.into_parts();
+                    let mut file = copy_body_to_file(file, body, format).await?;
+                    file.rewind().map_err(|e| {
+                        index_scheduler::Error::FileStore(file_store::Error::IoError(e))
+                    })?;
+                    Some(tempfile::NamedTempFile::from_parts(file, path))
+                }
+                None => None,
+            };
 
-    let temp_file = match tempfile() {
-        Ok(file) => file,
-        Err(e) => return Err(MeilisearchHttpError::Payload(ReceivePayload(Box::new(e)))),
+            let documents_count = tokio::task::spawn_blocking(move || {
+                let documents_count = file.as_ref().map_or(Ok(0), |ntf| {
+                    read_ndjson(ntf.as_file()).map_err(MeilisearchHttpError::DocumentFormat)
+                })?;
+
+                let update_file = file_store::File::from_parts(path, file);
+                update_file.persist()?;
+
+                Ok(documents_count)
+            })
+            .await?;
+
+            Ok(documents_count)
+        }
+        PayloadType::Json | PayloadType::Csv { delimiter: _ } => {
+            let temp_file = match tempfile() {
+                Ok(file) => file,
+                Err(e) => return Err(MeilisearchHttpError::Payload(ReceivePayload(Box::new(e)))),
+            };
+
+            let read_file = copy_body_to_file(temp_file, body, format).await?;
+            tokio::task::spawn_blocking(move || {
+                let documents_count = match format {
+                    PayloadType::Json => read_json(&read_file, &mut update_file)?,
+                    PayloadType::Csv { delimiter } => {
+                        read_csv(&read_file, &mut update_file, delimiter)?
+                    }
+                    PayloadType::Ndjson => {
+                        unreachable!("We already wrote the user content into the update file")
+                    }
+                };
+                // we NEED to persist the file here because we moved the `udpate_file` in another task.
+                update_file.persist()?;
+                Ok(documents_count)
+            })
+            .await
+        }
     };
 
-    let async_file = File::from_std(temp_file);
-    let mut buffer = BufWriter::new(async_file);
-
-    let mut buffer_write_size: usize = 0;
-    while let Some(result) = body.next().await {
-        let byte = result?;
-
-        if byte.is_empty() && buffer_write_size == 0 {
-            return Err(MeilisearchHttpError::MissingPayload(format));
-        }
-
-        match buffer.write_all(&byte).await {
-            Ok(()) => buffer_write_size += 1,
-            Err(e) => return Err(MeilisearchHttpError::Payload(ReceivePayload(Box::new(e)))),
-        }
-    }
-
-    if let Err(e) = buffer.flush().await {
-        return Err(MeilisearchHttpError::Payload(ReceivePayload(Box::new(e))));
-    }
-
-    if buffer_write_size == 0 {
-        return Err(MeilisearchHttpError::MissingPayload(format));
-    }
-
-    if let Err(e) = buffer.seek(std::io::SeekFrom::Start(0)).await {
-        return Err(MeilisearchHttpError::Payload(ReceivePayload(Box::new(e))));
-    }
-
-    let read_file = buffer.into_inner().into_std().await;
-    let documents_count = tokio::task::spawn_blocking(move || {
-        let documents_count = match format {
-            PayloadType::Json => read_json(&read_file, &mut update_file)?,
-            PayloadType::Csv { delimiter } => read_csv(&read_file, &mut update_file, delimiter)?,
-            PayloadType::Ndjson => read_ndjson(&read_file, &mut update_file)?,
-        };
-        // we NEED to persist the file here because we moved the `udpate_file` in another task.
-        update_file.persist()?;
-        Ok(documents_count)
-    })
-    .await;
-
     let documents_count = match documents_count {
         Ok(Ok(documents_count)) => documents_count,
         // in this case the file has not possibly be persisted.
@@ -703,6 +709,39 @@ async fn document_addition(
     Ok(task.into())
 }
 
+async fn copy_body_to_file(
+    output: std::fs::File,
+    mut body: Payload,
+    format: PayloadType,
+) -> Result<std::fs::File, MeilisearchHttpError> {
+    let async_file = File::from_std(output);
+    let mut buffer = BufWriter::new(async_file);
+    let mut buffer_write_size: usize = 0;
+    while let Some(result) = body.next().await {
+        let byte = result?;
+
+        if byte.is_empty() && buffer_write_size == 0 {
+            return Err(MeilisearchHttpError::MissingPayload(format));
+        }
+
+        match buffer.write_all(&byte).await {
+            Ok(()) => buffer_write_size += 1,
+            Err(e) => return Err(MeilisearchHttpError::Payload(ReceivePayload(Box::new(e)))),
+        }
+    }
+    if let Err(e) = buffer.flush().await {
+        return Err(MeilisearchHttpError::Payload(ReceivePayload(Box::new(e))));
+    }
+    if buffer_write_size == 0 {
+        return Err(MeilisearchHttpError::MissingPayload(format));
+    }
+    if let Err(e) = buffer.seek(std::io::SeekFrom::Start(0)).await {
+        return Err(MeilisearchHttpError::Payload(ReceivePayload(Box::new(e))));
+    }
+    let read_file = buffer.into_inner().into_std().await;
+    Ok(read_file)
+}
+
 pub async fn delete_documents_batch(
     index_scheduler: GuardedData<ActionPolicy<{ actions::DOCUMENTS_DELETE }>, Data<IndexScheduler>>,
     index_uid: web::Path<String>,

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

@@ -185,7 +185,8 @@ pub async fn search(
 
     let index = index_scheduler.index(&index_uid)?;
     let features = index_scheduler.features();
-    let search_kind = search_kind(&search_query, &index_scheduler, &index, features)?;
+    let search_kind =
+        search_kind(&search_query, &index_scheduler, index_uid.to_string(), &index, features)?;
     let permit = search_queue.try_get_search_permit().await?;
     let search_result = tokio::task::spawn_blocking(move || {
         perform_facet_search(

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

@@ -5,7 +5,7 @@ use actix_web::web::Data;
 use actix_web::{web, HttpRequest, HttpResponse};
 use deserr::actix_web::{AwebJson, AwebQueryParameter};
 use deserr::{DeserializeError, Deserr, ValuePointerRef};
-use index_scheduler::IndexScheduler;
+use index_scheduler::{Error, IndexScheduler};
 use meilisearch_types::deserr::query_params::Param;
 use meilisearch_types::deserr::{immutable_field_error, DeserrJsonError, DeserrQueryParamError};
 use meilisearch_types::error::deserr_codes::*;
@@ -107,7 +107,10 @@ pub async fn list_indexes(
             if !filters.is_index_authorized(uid) {
                 return Ok(None);
             }
-            Ok(Some(IndexView::new(uid.to_string(), index)?))
+            Ok(Some(
+                IndexView::new(uid.to_string(), index)
+                    .map_err(|e| Error::from_milli(e, Some(uid.to_string())))?,
+            ))
         })?;
     // Won't cause to open all indexes because IndexView doesn't keep the `Index` opened.
     let indexes: Vec<IndexView> = indexes.into_iter().flatten().collect();

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

@@ -243,11 +243,19 @@ pub async fn search_with_url_query(
     let index = index_scheduler.index(&index_uid)?;
     let features = index_scheduler.features();
 
-    let search_kind = search_kind(&query, index_scheduler.get_ref(), &index, features)?;
+    let search_kind =
+        search_kind(&query, index_scheduler.get_ref(), index_uid.to_string(), &index, features)?;
     let retrieve_vector = RetrieveVectors::new(query.retrieve_vectors, features)?;
     let permit = search_queue.try_get_search_permit().await?;
     let search_result = tokio::task::spawn_blocking(move || {
-        perform_search(&index, query, search_kind, retrieve_vector, index_scheduler.features())
+        perform_search(
+            index_uid.to_string(),
+            &index,
+            query,
+            search_kind,
+            retrieve_vector,
+            index_scheduler.features(),
+        )
     })
     .await;
     permit.drop().await;
@@ -287,12 +295,20 @@ pub async fn search_with_post(
 
     let features = index_scheduler.features();
 
-    let search_kind = search_kind(&query, index_scheduler.get_ref(), &index, features)?;
+    let search_kind =
+        search_kind(&query, index_scheduler.get_ref(), index_uid.to_string(), &index, features)?;
     let retrieve_vectors = RetrieveVectors::new(query.retrieve_vectors, features)?;
 
     let permit = search_queue.try_get_search_permit().await?;
     let search_result = tokio::task::spawn_blocking(move || {
-        perform_search(&index, query, search_kind, retrieve_vectors, index_scheduler.features())
+        perform_search(
+            index_uid.to_string(),
+            &index,
+            query,
+            search_kind,
+            retrieve_vectors,
+            index_scheduler.features(),
+        )
     })
     .await;
     permit.drop().await;
@@ -314,6 +330,7 @@ pub async fn search_with_post(
 pub fn search_kind(
     query: &SearchQuery,
     index_scheduler: &IndexScheduler,
+    index_uid: String,
     index: &milli::Index,
     features: RoFeatures,
 ) -> Result<SearchKind, ResponseError> {
@@ -332,7 +349,7 @@ pub fn search_kind(
         (None, _, None) => Ok(SearchKind::KeywordOnly),
         // hybrid.semantic_ratio == 1.0 => vector
         (_, Some(HybridQuery { semantic_ratio, embedder }), v) if **semantic_ratio == 1.0 => {
-            SearchKind::semantic(index_scheduler, index, embedder, v.map(|v| v.len()))
+            SearchKind::semantic(index_scheduler, index_uid, index, embedder, v.map(|v| v.len()))
         }
         // hybrid.semantic_ratio == 0.0 => keyword
         (_, Some(HybridQuery { semantic_ratio, embedder: _ }), _) if **semantic_ratio == 0.0 => {
@@ -340,13 +357,14 @@ pub fn search_kind(
         }
         // no query, hybrid, vector => semantic
         (None, Some(HybridQuery { semantic_ratio: _, embedder }), Some(v)) => {
-            SearchKind::semantic(index_scheduler, index, embedder, Some(v.len()))
+            SearchKind::semantic(index_scheduler, index_uid, index, embedder, Some(v.len()))
         }
         // query, no hybrid, no vector => keyword
         (Some(_), None, None) => Ok(SearchKind::KeywordOnly),
         // query, hybrid, maybe vector => hybrid
         (Some(_), Some(HybridQuery { semantic_ratio, embedder }), v) => SearchKind::hybrid(
             index_scheduler,
+            index_uid,
             index,
             embedder,
             **semantic_ratio,

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

@@ -17,6 +17,32 @@ use crate::extractors::authentication::GuardedData;
 use crate::routes::{get_task_id, is_dry_run, SummarizedTaskView};
 use crate::Opt;
 
+/// This macro generates the routes for the settings.
+///
+/// It takes a list of settings and generates a module for each setting.
+/// Each module contains the `get`, `update` and `delete` routes for the setting.
+///
+/// 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},)*) => {
+        $(
+            make_setting_route!($route, $update_verb, $type, $err_ty, $attr, $camelcase_attr, $analytics);
+        )*
+
+        pub fn configure(cfg: &mut web::ServiceConfig) {
+            use crate::extractors::sequential_extractor::SeqHandler;
+            cfg.service(
+                web::resource("")
+                .route(web::patch().to(SeqHandler(update_all)))
+                .route(web::get().to(SeqHandler(get_all)))
+                .route(web::delete().to(SeqHandler(delete_all))))
+                $(.service($attr::resources()))*;
+        }
+
+        pub const ALL_SETTINGS_NAMES: &[&str] = &[$(stringify!($attr)),*];
+    };
+}
+
 #[macro_export]
 macro_rules! make_setting_route {
     ($route:literal, $update_verb:ident, $type:ty, $err_ty:ty, $attr:ident, $camelcase_attr:literal, $analytics:ident) => {
@@ -153,279 +179,227 @@ macro_rules! make_setting_route {
     };
 }
 
-make_setting_route!(
-    "/filterable-attributes",
-    put,
-    std::collections::BTreeSet<String>,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsFilterableAttributes,
-    >,
-    filterable_attributes,
-    "filterableAttributes",
-    FilterableAttributesAnalytics
-);
-
-make_setting_route!(
-    "/sortable-attributes",
-    put,
-    std::collections::BTreeSet<String>,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsSortableAttributes,
-    >,
-    sortable_attributes,
-    "sortableAttributes",
-    SortableAttributesAnalytics
-);
-
-make_setting_route!(
-    "/displayed-attributes",
-    put,
-    Vec<String>,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsDisplayedAttributes,
-    >,
-    displayed_attributes,
-    "displayedAttributes",
-    DisplayedAttributesAnalytics
-);
-
-make_setting_route!(
-    "/typo-tolerance",
-    patch,
-    meilisearch_types::settings::TypoSettings,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsTypoTolerance,
-    >,
-    typo_tolerance,
-    "typoTolerance",
-    TypoToleranceAnalytics
-);
-
-make_setting_route!(
-    "/searchable-attributes",
-    put,
-    Vec<String>,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsSearchableAttributes,
-    >,
-    searchable_attributes,
-    "searchableAttributes",
-    SearchableAttributesAnalytics
-);
-
-make_setting_route!(
-    "/stop-words",
-    put,
-    std::collections::BTreeSet<String>,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsStopWords,
-    >,
-    stop_words,
-    "stopWords",
-    StopWordsAnalytics
-);
-
-make_setting_route!(
-    "/non-separator-tokens",
-    put,
-    std::collections::BTreeSet<String>,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsNonSeparatorTokens,
-    >,
-    non_separator_tokens,
-    "nonSeparatorTokens",
-    NonSeparatorTokensAnalytics
-);
-
-make_setting_route!(
-    "/separator-tokens",
-    put,
-    std::collections::BTreeSet<String>,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsSeparatorTokens,
-    >,
-    separator_tokens,
-    "separatorTokens",
-    SeparatorTokensAnalytics
-);
-
-make_setting_route!(
-    "/dictionary",
-    put,
-    std::collections::BTreeSet<String>,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsDictionary,
-    >,
-    dictionary,
-    "dictionary",
-    DictionaryAnalytics
-);
-
-make_setting_route!(
-    "/synonyms",
-    put,
-    std::collections::BTreeMap<String, Vec<String>>,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsSynonyms,
-    >,
-    synonyms,
-    "synonyms",
-    SynonymsAnalytics
-);
-
-make_setting_route!(
-    "/distinct-attribute",
-    put,
-    String,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsDistinctAttribute,
-    >,
-    distinct_attribute,
-    "distinctAttribute",
-    DistinctAttributeAnalytics
-);
-
-make_setting_route!(
-    "/proximity-precision",
-    put,
-    meilisearch_types::settings::ProximityPrecisionView,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsProximityPrecision,
-    >,
-    proximity_precision,
-    "proximityPrecision",
-    ProximityPrecisionAnalytics
-);
-
-make_setting_route!(
-    "/localized-attributes",
-    put,
-    Vec<meilisearch_types::locales::LocalizedAttributesRuleView>,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsLocalizedAttributes,
-    >,
-    localized_attributes,
-    "localizedAttributes",
-    LocalesAnalytics
-);
-
-make_setting_route!(
-    "/ranking-rules",
-    put,
-    Vec<meilisearch_types::settings::RankingRuleView>,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsRankingRules,
-    >,
-    ranking_rules,
-    "rankingRules",
-    RankingRulesAnalytics
-);
-
-make_setting_route!(
-    "/faceting",
-    patch,
-    meilisearch_types::settings::FacetingSettings,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsFaceting,
-    >,
-    faceting,
-    "faceting",
-    FacetingAnalytics
-);
-
-make_setting_route!(
-    "/pagination",
-    patch,
-    meilisearch_types::settings::PaginationSettings,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsPagination,
-    >,
-    pagination,
-    "pagination",
-    PaginationAnalytics
-);
-
-make_setting_route!(
-    "/embedders",
-    patch,
-    std::collections::BTreeMap<String, Setting<meilisearch_types::milli::vector::settings::EmbeddingSettings>>,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsEmbedders,
-    >,
-    embedders,
-    "embedders",
-    EmbeddersAnalytics
-);
-
-make_setting_route!(
-    "/search-cutoff-ms",
-    put,
-    u64,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsSearchCutoffMs,
-    >,
-    search_cutoff_ms,
-    "searchCutoffMs",
-    SearchCutoffMsAnalytics
-);
-
-make_setting_route!(
-    "/facet-search",
-    put,
-    bool,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsFacetSearch,
-    >,
-    facet_search,
-    "facetSearch",
-    FacetSearchAnalytics
-);
-
-make_setting_route!(
-    "/prefix-search",
-    put,
-    meilisearch_types::settings::PrefixSearchSettings,
-    meilisearch_types::deserr::DeserrJsonError<
-        meilisearch_types::error::deserr_codes::InvalidSettingsPrefixSearch,
-    >,
-    prefix_search,
-    "prefixSearch",
-    PrefixSearchAnalytics
-);
-
-macro_rules! generate_configure {
-    ($($mod:ident),*) => {
-        pub fn configure(cfg: &mut web::ServiceConfig) {
-            use crate::extractors::sequential_extractor::SeqHandler;
-            cfg.service(
-                web::resource("")
-                .route(web::patch().to(SeqHandler(update_all)))
-                .route(web::get().to(SeqHandler(get_all)))
-                .route(web::delete().to(SeqHandler(delete_all))))
-                $(.service($mod::resources()))*;
-        }
-    };
-}
-
-generate_configure!(
-    filterable_attributes,
-    sortable_attributes,
-    displayed_attributes,
-    localized_attributes,
-    searchable_attributes,
-    distinct_attribute,
-    proximity_precision,
-    stop_words,
-    separator_tokens,
-    non_separator_tokens,
-    dictionary,
-    synonyms,
-    ranking_rules,
-    typo_tolerance,
-    pagination,
-    faceting,
-    embedders,
-    search_cutoff_ms
+make_setting_routes!(
+    {
+        route: "/filterable-attributes",
+        update_verb: put,
+        value_type: std::collections::BTreeSet<String>,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsFilterableAttributes,
+        >,
+        attr: filterable_attributes,
+        camelcase_attr: "filterableAttributes",
+        analytics: FilterableAttributesAnalytics
+    },
+    {
+        route: "/sortable-attributes",
+        update_verb: put,
+        value_type: std::collections::BTreeSet<String>,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsSortableAttributes,
+        >,
+        attr: sortable_attributes,
+        camelcase_attr: "sortableAttributes",
+        analytics: SortableAttributesAnalytics
+    },
+    {
+        route: "/displayed-attributes",
+        update_verb: put,
+        value_type: Vec<String>,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsDisplayedAttributes,
+        >,
+        attr: displayed_attributes,
+        camelcase_attr: "displayedAttributes",
+        analytics: DisplayedAttributesAnalytics
+    },
+    {
+        route: "/typo-tolerance",
+        update_verb: patch,
+        value_type: meilisearch_types::settings::TypoSettings,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsTypoTolerance,
+        >,
+        attr: typo_tolerance,
+        camelcase_attr: "typoTolerance",
+        analytics: TypoToleranceAnalytics
+    },
+    {
+        route: "/searchable-attributes",
+        update_verb: put,
+        value_type: Vec<String>,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsSearchableAttributes,
+        >,
+        attr: searchable_attributes,
+        camelcase_attr: "searchableAttributes",
+        analytics: SearchableAttributesAnalytics
+    },
+    {
+        route: "/stop-words",
+        update_verb: put,
+        value_type: std::collections::BTreeSet<String>,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsStopWords,
+        >,
+        attr: stop_words,
+        camelcase_attr: "stopWords",
+        analytics: StopWordsAnalytics
+    },
+    {
+        route: "/non-separator-tokens",
+        update_verb: put,
+        value_type: std::collections::BTreeSet<String>,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsNonSeparatorTokens,
+        >,
+        attr: non_separator_tokens,
+        camelcase_attr: "nonSeparatorTokens",
+        analytics: NonSeparatorTokensAnalytics
+    },
+    {
+        route: "/separator-tokens",
+        update_verb: put,
+        value_type: std::collections::BTreeSet<String>,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsSeparatorTokens,
+        >,
+        attr: separator_tokens,
+        camelcase_attr: "separatorTokens",
+        analytics: SeparatorTokensAnalytics
+    },
+    {
+        route: "/dictionary",
+        update_verb: put,
+        value_type: std::collections::BTreeSet<String>,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsDictionary,
+        >,
+        attr: dictionary,
+        camelcase_attr: "dictionary",
+        analytics: DictionaryAnalytics
+    },
+    {
+        route: "/synonyms",
+        update_verb: put,
+        value_type: std::collections::BTreeMap<String, Vec<String>>,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsSynonyms,
+        >,
+        attr: synonyms,
+        camelcase_attr: "synonyms",
+        analytics: SynonymsAnalytics
+    },
+    {
+        route: "/distinct-attribute",
+        update_verb: put,
+        value_type: String,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsDistinctAttribute,
+        >,
+        attr: distinct_attribute,
+        camelcase_attr: "distinctAttribute",
+        analytics: DistinctAttributeAnalytics
+    },
+    {
+        route: "/proximity-precision",
+        update_verb: put,
+        value_type: meilisearch_types::settings::ProximityPrecisionView,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsProximityPrecision,
+        >,
+        attr: proximity_precision,
+        camelcase_attr: "proximityPrecision",
+        analytics: ProximityPrecisionAnalytics
+    },
+    {
+        route: "/localized-attributes",
+        update_verb: put,
+        value_type: Vec<meilisearch_types::locales::LocalizedAttributesRuleView>,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsLocalizedAttributes,
+        >,
+        attr: localized_attributes,
+        camelcase_attr: "localizedAttributes",
+        analytics: LocalesAnalytics
+    },
+    {
+        route: "/ranking-rules",
+        update_verb: put,
+        value_type: Vec<meilisearch_types::settings::RankingRuleView>,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsRankingRules,
+        >,
+        attr: ranking_rules,
+        camelcase_attr: "rankingRules",
+        analytics: RankingRulesAnalytics
+    },
+    {
+        route: "/faceting",
+        update_verb: patch,
+        value_type: meilisearch_types::settings::FacetingSettings,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsFaceting,
+        >,
+        attr: faceting,
+        camelcase_attr: "faceting",
+        analytics: FacetingAnalytics
+    },
+    {
+        route: "/pagination",
+        update_verb: patch,
+        value_type: meilisearch_types::settings::PaginationSettings,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsPagination,
+        >,
+        attr: pagination,
+        camelcase_attr: "pagination",
+        analytics: PaginationAnalytics
+    },
+    {
+        route: "/embedders",
+        update_verb: patch,
+        value_type: std::collections::BTreeMap<String, Setting<meilisearch_types::milli::vector::settings::EmbeddingSettings>>,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsEmbedders,
+        >,
+        attr: embedders,
+        camelcase_attr: "embedders",
+        analytics: EmbeddersAnalytics
+    },
+    {
+        route: "/search-cutoff-ms",
+        update_verb: put,
+        value_type: u64,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsSearchCutoffMs,
+        >,
+        attr: search_cutoff_ms,
+        camelcase_attr: "searchCutoffMs",
+        analytics: SearchCutoffMsAnalytics
+    },
+    {
+        route: "/facet-search",
+        update_verb: put,
+        value_type: bool,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsFacetSearch,
+        >,
+        attr: facet_search,
+        camelcase_attr: "facetSearch",
+        analytics: FacetSearchAnalytics
+    },
+    {
+        route: "/prefix-search",
+        update_verb: put,
+        value_type: meilisearch_types::settings::PrefixSearchSettings,
+        err_type: meilisearch_types::deserr::DeserrJsonError<
+            meilisearch_types::error::deserr_codes::InvalidSettingsPrefixSearch,
+        >,
+        attr: prefix_search,
+        camelcase_attr: "prefixSearch",
+        analytics: PrefixSearchAnalytics
+    },
 );
 
 pub async fn update_all(

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

@@ -103,8 +103,13 @@ async fn similar(
 
     let index = index_scheduler.index(&index_uid)?;
 
-    let (embedder_name, embedder, quantized) =
-        SearchKind::embedder(&index_scheduler, &index, &query.embedder, None)?;
+    let (embedder_name, embedder, quantized) = SearchKind::embedder(
+        &index_scheduler,
+        index_uid.to_string(),
+        &index,
+        &query.embedder,
+        None,
+    )?;
 
     tokio::task::spawn_blocking(move || {
         perform_similar(

crates/meilisearch/src/routes/multi_search.rs

@@ -125,14 +125,28 @@ pub async fn multi_search_with_post(
                         })
                         .with_index(query_index)?;
 
-                    let search_kind =
-                        search_kind(&query, index_scheduler.get_ref(), &index, features)
-                            .with_index(query_index)?;
+                    let index_uid_str = index_uid.to_string();
+
+                    let search_kind = search_kind(
+                        &query,
+                        index_scheduler.get_ref(),
+                        index_uid_str.clone(),
+                        &index,
+                        features,
+                    )
+                    .with_index(query_index)?;
                     let retrieve_vector = RetrieveVectors::new(query.retrieve_vectors, features)
                         .with_index(query_index)?;
 
                     let search_result = tokio::task::spawn_blocking(move || {
-                        perform_search(&index, query, search_kind, retrieve_vector, features)
+                        perform_search(
+                            index_uid_str.clone(),
+                            &index,
+                            query,
+                            search_kind,
+                            retrieve_vector,
+                            features,
+                        )
                     })
                     .await
                     .with_index(query_index)?;

crates/meilisearch/src/search/federated.rs

@@ -560,7 +560,8 @@ pub fn perform_federated_search(
             // use an immediately invoked lambda to capture the result without returning from the function
 
             let res: Result<(), ResponseError> = (|| {
-                let search_kind = search_kind(&query, index_scheduler, &index, features)?;
+                let search_kind =
+                    search_kind(&query, index_scheduler, index_uid.to_string(), &index, features)?;
 
                 let canonicalization_kind = match (&search_kind, &query.q) {
                     (SearchKind::SemanticOnly { .. }, _) => {
@@ -636,7 +637,8 @@ pub fn perform_federated_search(
                 search.offset(0);
                 search.limit(required_hit_count);
 
-                let (result, _semantic_hit_count) = super::search_from_kind(search_kind, search)?;
+                let (result, _semantic_hit_count) =
+                    super::search_from_kind(index_uid.to_string(), search_kind, search)?;
                 let format = AttributesFormat {
                     attributes_to_retrieve: query.attributes_to_retrieve,
                     retrieve_vectors,
@@ -670,7 +672,10 @@ pub fn perform_federated_search(
 
                 let formatter_builder = HitMaker::formatter_builder(matching_words, tokenizer);
 
-                let hit_maker = HitMaker::new(&index, &rtxn, format, formatter_builder)?;
+                let hit_maker =
+                    HitMaker::new(&index, &rtxn, format, formatter_builder).map_err(|e| {
+                        MeilisearchHttpError::from_milli(e, Some(index_uid.to_string()))
+                    })?;
 
                 results_by_query.push(SearchResultByQuery {
                     federation_options,

crates/meilisearch/src/search/mod.rs

@@ -19,7 +19,9 @@ use meilisearch_types::locales::Locale;
 use meilisearch_types::milli::score_details::{ScoreDetails, ScoringStrategy};
 use meilisearch_types::milli::vector::parsed_vectors::ExplicitVectors;
 use meilisearch_types::milli::vector::Embedder;
-use meilisearch_types::milli::{FacetValueHit, OrderBy, SearchForFacetValues, TimeBudget};
+use meilisearch_types::milli::{
+    FacetValueHit, InternalError, OrderBy, SearchForFacetValues, TimeBudget,
+};
 use meilisearch_types::settings::DEFAULT_PAGINATION_MAX_TOTAL_HITS;
 use meilisearch_types::{milli, Document};
 use milli::tokenizer::{Language, TokenizerBuilder};
@@ -281,35 +283,38 @@ pub enum SearchKind {
 impl SearchKind {
     pub(crate) fn semantic(
         index_scheduler: &index_scheduler::IndexScheduler,
+        index_uid: String,
         index: &Index,
         embedder_name: &str,
         vector_len: Option<usize>,
     ) -> Result<Self, ResponseError> {
         let (embedder_name, embedder, quantized) =
-            Self::embedder(index_scheduler, index, embedder_name, vector_len)?;
+            Self::embedder(index_scheduler, index_uid, index, embedder_name, vector_len)?;
         Ok(Self::SemanticOnly { embedder_name, embedder, quantized })
     }
 
     pub(crate) fn hybrid(
         index_scheduler: &index_scheduler::IndexScheduler,
+        index_uid: String,
         index: &Index,
         embedder_name: &str,
         semantic_ratio: f32,
         vector_len: Option<usize>,
     ) -> Result<Self, ResponseError> {
         let (embedder_name, embedder, quantized) =
-            Self::embedder(index_scheduler, index, embedder_name, vector_len)?;
+            Self::embedder(index_scheduler, index_uid, index, embedder_name, vector_len)?;
         Ok(Self::Hybrid { embedder_name, embedder, quantized, semantic_ratio })
     }
 
     pub(crate) fn embedder(
         index_scheduler: &index_scheduler::IndexScheduler,
+        index_uid: String,
         index: &Index,
         embedder_name: &str,
         vector_len: Option<usize>,
     ) -> Result<(String, Arc<Embedder>, bool), ResponseError> {
         let embedder_configs = index.embedding_configs(&index.read_txn()?)?;
-        let embedders = index_scheduler.embedders(embedder_configs)?;
+        let embedders = index_scheduler.embedders(index_uid, embedder_configs)?;
 
         let (embedder, _, quantized) = embedders
             .get(embedder_name)
@@ -890,6 +895,7 @@ fn prepare_search<'t>(
 }
 
 pub fn perform_search(
+    index_uid: String,
     index: &Index,
     query: SearchQuery,
     search_kind: SearchKind,
@@ -916,7 +922,7 @@ pub fn perform_search(
             used_negative_operator,
         },
         semantic_hit_count,
-    ) = search_from_kind(search_kind, search)?;
+    ) = search_from_kind(index_uid, search_kind, search)?;
 
     let SearchQuery {
         q,
@@ -1069,17 +1075,27 @@ fn compute_facet_distribution_stats<S: AsRef<str>>(
 }
 
 pub fn search_from_kind(
+    index_uid: String,
     search_kind: SearchKind,
     search: milli::Search<'_>,
 ) -> Result<(milli::SearchResult, Option<u32>), MeilisearchHttpError> {
     let (milli_result, semantic_hit_count) = match &search_kind {
-        SearchKind::KeywordOnly => (search.execute()?, None),
+        SearchKind::KeywordOnly => {
+            let results = search
+                .execute()
+                .map_err(|e| MeilisearchHttpError::from_milli(e, Some(index_uid.to_string())))?;
+            (results, None)
+        }
         SearchKind::SemanticOnly { .. } => {
-            let results = search.execute()?;
+            let results = search
+                .execute()
+                .map_err(|e| MeilisearchHttpError::from_milli(e, Some(index_uid.to_string())))?;
             let semantic_hit_count = results.document_scores.len() as u32;
             (results, Some(semantic_hit_count))
         }
-        SearchKind::Hybrid { semantic_ratio, .. } => search.execute_hybrid(*semantic_ratio)?,
+        SearchKind::Hybrid { semantic_ratio, .. } => search
+            .execute_hybrid(*semantic_ratio)
+            .map_err(|e| MeilisearchHttpError::from_milli(e, Some(index_uid)))?,
     };
     Ok((milli_result, semantic_hit_count))
 }
@@ -1181,7 +1197,7 @@ impl<'a> HitMaker<'a> {
         rtxn: &'a RoTxn<'a>,
         format: AttributesFormat,
         mut formatter_builder: MatcherBuilder<'a>,
-    ) -> Result<Self, MeilisearchHttpError> {
+    ) -> milli::Result<Self> {
         formatter_builder.crop_marker(format.crop_marker);
         formatter_builder.highlight_prefix(format.highlight_pre_tag);
         formatter_builder.highlight_suffix(format.highlight_post_tag);
@@ -1276,11 +1292,7 @@ impl<'a> HitMaker<'a> {
         })
     }
 
-    pub fn make_hit(
-        &self,
-        id: u32,
-        score: &[ScoreDetails],
-    ) -> Result<SearchHit, MeilisearchHttpError> {
+    pub fn make_hit(&self, id: u32, score: &[ScoreDetails]) -> milli::Result<SearchHit> {
         let (_, obkv) =
             self.index.iter_documents(self.rtxn, std::iter::once(id))?.next().unwrap()?;
 
@@ -1323,7 +1335,10 @@ impl<'a> HitMaker<'a> {
                     .is_some_and(|conf| conf.user_provided.contains(id));
                 let embeddings =
                     ExplicitVectors { embeddings: Some(vector.into()), regenerate: !user_provided };
-                vectors.insert(name, serde_json::to_value(embeddings)?);
+                vectors.insert(
+                    name,
+                    serde_json::to_value(embeddings).map_err(InternalError::SerdeJson).unwrap(),
+                );
             }
             document.insert("_vectors".into(), vectors.into());
         }
@@ -1369,7 +1384,7 @@ fn make_hits<'a>(
     format: AttributesFormat,
     matching_words: milli::MatchingWords,
     documents_ids_scores: impl Iterator<Item = (u32, &'a Vec<ScoreDetails>)> + 'a,
-) -> Result<Vec<SearchHit>, MeilisearchHttpError> {
+) -> milli::Result<Vec<SearchHit>> {
     let mut documents = Vec::new();
 
     let dictionary = index.dictionary(rtxn)?;
@@ -1407,6 +1422,13 @@ pub fn perform_facet_search(
         None => TimeBudget::default(),
     };
 
+    if !index.facet_search(&rtxn)? {
+        return Err(ResponseError::from_msg(
+            "The facet search is disabled for this index".to_string(),
+            Code::FacetSearchDisabled,
+        ));
+    }
+
     // In the faceted search context, we want to use the intersection between the locales provided by the user
     // and the locales of the facet string.
     // If the facet string is not localized, we **ignore** the locales provided by the user because the facet data has no locale.
@@ -1690,12 +1712,12 @@ fn make_document(
     displayed_attributes: &BTreeSet<FieldId>,
     field_ids_map: &FieldsIdsMap,
     obkv: &obkv::KvReaderU16,
-) -> Result<Document, MeilisearchHttpError> {
+) -> milli::Result<Document> {
     let mut document = serde_json::Map::new();
 
     // recreate the original json
     for (key, value) in obkv.iter() {
-        let value = serde_json::from_slice(value)?;
+        let value = serde_json::from_slice(value).map_err(InternalError::SerdeJson).unwrap();
         let key = field_ids_map.name(key).expect("Missing field name").to_string();
 
         document.insert(key, value);
@@ -1720,7 +1742,7 @@ fn format_fields(
     displayable_ids: &BTreeSet<FieldId>,
     locales: Option<&[Language]>,
     localized_attributes: &[LocalizedAttributesRule],
-) -> Result<(Option<MatchesPosition>, Document), MeilisearchHttpError> {
+) -> milli::Result<(Option<MatchesPosition>, Document)> {
     let mut matches_position = compute_matches.then(BTreeMap::new);
     let mut document = document.clone();
 
@@ -1898,7 +1920,7 @@ fn parse_filter_array(arr: &[Value]) -> Result<Option<Filter>, MeilisearchHttpEr
         }
     }
 
-    Ok(Filter::from_array(ands)?)
+    Filter::from_array(ands).map_err(|e| MeilisearchHttpError::from_milli(e, None))
 }
 
 #[cfg(test)]

crates/meilisearch/tests/batches/mod.rs

@@ -224,7 +224,7 @@ async fn list_batches_status_and_type_filtered() {
 }
 
 #[actix_rt::test]
-async fn get_batch_filter_error() {
+async fn list_batch_filter_error() {
     let server = Server::new().await;
 
     let (response, code) = server.batches_filter("lol=pied").await;
@@ -284,6 +284,7 @@ async fn test_summarized_document_addition_or_update() {
         @r#"
     {
       "uid": 0,
+      "progress": null,
       "details": {
         "receivedDocuments": 1,
         "indexedDocuments": 1
@@ -314,6 +315,7 @@ async fn test_summarized_document_addition_or_update() {
         @r#"
     {
       "uid": 1,
+      "progress": null,
       "details": {
         "receivedDocuments": 1,
         "indexedDocuments": 1
@@ -349,6 +351,7 @@ async fn test_summarized_delete_documents_by_batch() {
         @r#"
     {
       "uid": 0,
+      "progress": null,
       "details": {
         "providedIds": 3,
         "deletedDocuments": 0
@@ -380,6 +383,7 @@ async fn test_summarized_delete_documents_by_batch() {
         @r#"
     {
       "uid": 2,
+      "progress": null,
       "details": {
         "providedIds": 1,
         "deletedDocuments": 0
@@ -416,6 +420,7 @@ async fn test_summarized_delete_documents_by_filter() {
         @r#"
     {
       "uid": 0,
+      "progress": null,
       "details": {
         "providedIds": 0,
         "deletedDocuments": 0,
@@ -448,6 +453,7 @@ async fn test_summarized_delete_documents_by_filter() {
         @r#"
     {
       "uid": 2,
+      "progress": null,
       "details": {
         "providedIds": 0,
         "deletedDocuments": 0,
@@ -480,6 +486,7 @@ async fn test_summarized_delete_documents_by_filter() {
         @r#"
     {
       "uid": 4,
+      "progress": null,
       "details": {
         "providedIds": 0,
         "deletedDocuments": 0,
@@ -516,6 +523,7 @@ async fn test_summarized_delete_document_by_id() {
         @r#"
     {
       "uid": 0,
+      "progress": null,
       "details": {
         "providedIds": 1,
         "deletedDocuments": 0
@@ -547,6 +555,7 @@ async fn test_summarized_delete_document_by_id() {
         @r#"
     {
       "uid": 2,
+      "progress": null,
       "details": {
         "providedIds": 1,
         "deletedDocuments": 0
@@ -594,6 +603,7 @@ async fn test_summarized_settings_update() {
         @r#"
     {
       "uid": 0,
+      "progress": null,
       "details": {
         "displayedAttributes": [
           "doggos",
@@ -638,6 +648,7 @@ async fn test_summarized_index_creation() {
         @r#"
     {
       "uid": 0,
+      "progress": null,
       "details": {},
       "stats": {
         "totalNbTasks": 1,
@@ -665,6 +676,7 @@ async fn test_summarized_index_creation() {
         @r#"
     {
       "uid": 1,
+      "progress": null,
       "details": {
         "primaryKey": "doggos"
       },
@@ -809,6 +821,7 @@ async fn test_summarized_index_update() {
         @r#"
     {
       "uid": 0,
+      "progress": null,
       "details": {},
       "stats": {
         "totalNbTasks": 1,
@@ -836,6 +849,7 @@ async fn test_summarized_index_update() {
         @r#"
     {
       "uid": 1,
+      "progress": null,
       "details": {
         "primaryKey": "bones"
       },
@@ -868,6 +882,7 @@ async fn test_summarized_index_update() {
         @r#"
     {
       "uid": 3,
+      "progress": null,
       "details": {},
       "stats": {
         "totalNbTasks": 1,
@@ -895,6 +910,7 @@ async fn test_summarized_index_update() {
         @r#"
     {
       "uid": 4,
+      "progress": null,
       "details": {
         "primaryKey": "bones"
       },
@@ -932,6 +948,7 @@ async fn test_summarized_index_swap() {
         @r#"
     {
       "uid": 0,
+      "progress": null,
       "details": {
         "swaps": [
           {
@@ -972,6 +989,7 @@ async fn test_summarized_index_swap() {
         @r#"
     {
       "uid": 3,
+      "progress": null,
       "details": {
         "swaps": [
           {
@@ -1014,6 +1032,7 @@ async fn test_summarized_batch_cancelation() {
         @r#"
     {
       "uid": 1,
+      "progress": null,
       "details": {
         "matchedTasks": 1,
         "canceledTasks": 0,
@@ -1051,6 +1070,7 @@ async fn test_summarized_batch_deletion() {
         @r#"
     {
       "uid": 1,
+      "progress": null,
       "details": {
         "matchedTasks": 1,
         "deletedTasks": 1,
@@ -1084,6 +1104,7 @@ async fn test_summarized_dump_creation() {
         @r#"
     {
       "uid": 0,
+      "progress": null,
       "details": {
         "dumpUid": "[dumpUid]"
       },

crates/meilisearch/tests/common/mod.rs

@@ -52,6 +52,25 @@ impl Value {
         }
         self
     }
+
+    /// Return `true` if the `status` field is set to `failed`.
+    /// Panic if the `status` field doesn't exists.
+    #[track_caller]
+    pub fn is_fail(&self) -> bool {
+        if !self["status"].is_string() {
+            panic!("Called `is_fail` on {}", serde_json::to_string_pretty(&self.0).unwrap());
+        }
+        self["status"] == serde_json::Value::String(String::from("failed"))
+    }
+
+    // Panic if the json doesn't contain the `status` field set to "succeeded"
+    #[track_caller]
+    pub fn failed(&self) -> &Self {
+        if !self.is_fail() {
+            panic!("Called failed on {}", serde_json::to_string_pretty(&self.0).unwrap());
+        }
+        self
+    }
 }
 
 impl From<serde_json::Value> for Value {

crates/meilisearch/tests/documents/add_documents.rs

@@ -1220,9 +1220,89 @@ async fn replace_document() {
 #[actix_rt::test]
 async fn add_no_documents() {
     let server = Server::new().await;
-    let index = server.index("test");
-    let (_response, code) = index.add_documents(json!([]), None).await;
+    let index = server.index("kefir");
+    let (task, code) = index.add_documents(json!([]), None).await;
     snapshot!(code, @"202 Accepted");
+    let task = server.wait_task(task.uid()).await;
+    let task = task.succeeded();
+    snapshot!(task, @r#"
+    {
+      "uid": "[uid]",
+      "batchUid": "[batch_uid]",
+      "indexUid": "kefir",
+      "status": "succeeded",
+      "type": "documentAdditionOrUpdate",
+      "canceledBy": null,
+      "details": {
+        "receivedDocuments": 0,
+        "indexedDocuments": 0
+      },
+      "error": null,
+      "duration": "[duration]",
+      "enqueuedAt": "[date]",
+      "startedAt": "[date]",
+      "finishedAt": "[date]"
+    }
+    "#);
+
+    let (task, _code) = index.add_documents(json!([]), Some("kefkef")).await;
+    let task = server.wait_task(task.uid()).await;
+    let task = task.succeeded();
+    snapshot!(task, @r#"
+    {
+      "uid": "[uid]",
+      "batchUid": "[batch_uid]",
+      "indexUid": "kefir",
+      "status": "succeeded",
+      "type": "documentAdditionOrUpdate",
+      "canceledBy": null,
+      "details": {
+        "receivedDocuments": 0,
+        "indexedDocuments": 0
+      },
+      "error": null,
+      "duration": "[duration]",
+      "enqueuedAt": "[date]",
+      "startedAt": "[date]",
+      "finishedAt": "[date]"
+    }
+    "#);
+
+    let (task, _code) = index.add_documents(json!([{ "kefkef": 1 }]), None).await;
+    let task = server.wait_task(task.uid()).await;
+    let task = task.succeeded();
+    snapshot!(task, @r#"
+    {
+      "uid": "[uid]",
+      "batchUid": "[batch_uid]",
+      "indexUid": "kefir",
+      "status": "succeeded",
+      "type": "documentAdditionOrUpdate",
+      "canceledBy": null,
+      "details": {
+        "receivedDocuments": 1,
+        "indexedDocuments": 1
+      },
+      "error": null,
+      "duration": "[duration]",
+      "enqueuedAt": "[date]",
+      "startedAt": "[date]",
+      "finishedAt": "[date]"
+    }
+    "#);
+    let (documents, _status) = index.get_all_documents(GetAllDocumentsOptions::default()).await;
+    snapshot!(documents, @r#"
+    {
+      "results": [
+        {
+          "kefkef": 1
+        }
+      ],
+      "offset": 0,
+      "limit": 20,
+      "total": 1
+    }
+    "#);
 }
 
 #[actix_rt::test]
@@ -1264,15 +1344,18 @@ async fn error_add_documents_bad_document_id() {
     let server = Server::new().await;
     let index = server.index("test");
     index.create(Some("docid")).await;
+
+    // unsupported characters
+
     let documents = json!([
         {
             "docid": "foo & bar",
             "content": "foobar"
         }
     ]);
-    index.add_documents(documents, None).await;
-    index.wait_task(1).await;
-    let (response, code) = index.get_task(1).await;
+    let (value, _code) = index.add_documents(documents, None).await;
+    index.wait_task(value.uid()).await;
+    let (response, code) = index.get_task(value.uid()).await;
     snapshot!(code, @"200 OK");
     snapshot!(json_string!(response, { ".duration" => "[duration]", ".enqueuedAt" => "[date]", ".startedAt" => "[date]", ".finishedAt" => "[date]" }),
         @r###"
@@ -1288,7 +1371,81 @@ async fn error_add_documents_bad_document_id() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Document identifier `\"foo & bar\"` is invalid. A document identifier can be of type integer or string, only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_), and can not be more than 512 bytes.",
+        "message": "Document identifier `\"foo & bar\"` is invalid. A document identifier can be of type integer or string, only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_), and can not be more than 511 bytes.",
+        "code": "invalid_document_id",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_document_id"
+      },
+      "duration": "[duration]",
+      "enqueuedAt": "[date]",
+      "startedAt": "[date]",
+      "finishedAt": "[date]"
+    }
+    "###);
+
+    // More than 512 bytes
+    let documents = json!([
+        {
+            "docid": "a".repeat(600),
+            "content": "foobar"
+        }
+    ]);
+    let (value, _code) = index.add_documents(documents, None).await;
+    index.wait_task(value.uid()).await;
+    let (response, code) = index.get_task(value.uid()).await;
+    snapshot!(code, @"200 OK");
+    snapshot!(json_string!(response, { ".duration" => "[duration]", ".enqueuedAt" => "[date]", ".startedAt" => "[date]", ".finishedAt" => "[date]" }),
+      @r###"
+    {
+      "uid": 2,
+      "batchUid": 2,
+      "indexUid": "test",
+      "status": "failed",
+      "type": "documentAdditionOrUpdate",
+      "canceledBy": null,
+      "details": {
+        "receivedDocuments": 1,
+        "indexedDocuments": 0
+      },
+      "error": {
+        "message": "Document identifier `\"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa\"` is invalid. A document identifier can be of type integer or string, only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_), and can not be more than 511 bytes.",
+        "code": "invalid_document_id",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_document_id"
+      },
+      "duration": "[duration]",
+      "enqueuedAt": "[date]",
+      "startedAt": "[date]",
+      "finishedAt": "[date]"
+    }
+    "###);
+
+    // Exactly 512 bytes
+    let documents = json!([
+        {
+            "docid": "a".repeat(512),
+            "content": "foobar"
+        }
+    ]);
+    let (value, _code) = index.add_documents(documents, None).await;
+    index.wait_task(value.uid()).await;
+    let (response, code) = index.get_task(value.uid()).await;
+    snapshot!(code, @"200 OK");
+    snapshot!(json_string!(response, { ".duration" => "[duration]", ".enqueuedAt" => "[date]", ".startedAt" => "[date]", ".finishedAt" => "[date]" }),
+    @r###"
+    {
+      "uid": 3,
+      "batchUid": 3,
+      "indexUid": "test",
+      "status": "failed",
+      "type": "documentAdditionOrUpdate",
+      "canceledBy": null,
+      "details": {
+        "receivedDocuments": 1,
+        "indexedDocuments": 0
+      },
+      "error": {
+        "message": "Document identifier `\"aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa\"` is invalid. A document identifier can be of type integer or string, only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_), and can not be more than 511 bytes.",
         "code": "invalid_document_id",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_id"
@@ -1681,7 +1838,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "The `_geo` field in the document with the id: `\"11\"` is not an object. Was expecting an object with the `_geo.lat` and `_geo.lng` fields but instead got `\"foobar\"`.",
+        "message": "Index `test`: The `_geo` field in the document with the id: `\"11\"` is not an object. Was expecting an object with the `_geo.lat` and `_geo.lng` fields but instead got `\"foobar\"`.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1719,7 +1876,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Could not find latitude nor longitude in the document with the id: `\"11\"`. Was expecting `_geo.lat` and `_geo.lng` fields.",
+        "message": "Index `test`: Could not find latitude nor longitude in the document with the id: `\"11\"`. Was expecting `_geo.lat` and `_geo.lng` fields.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1757,7 +1914,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Could not find latitude nor longitude in the document with the id: `\"11\"`. Was expecting `_geo.lat` and `_geo.lng` fields.",
+        "message": "Index `test`: Could not find latitude nor longitude in the document with the id: `\"11\"`. Was expecting `_geo.lat` and `_geo.lng` fields.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1795,7 +1952,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Could not find longitude in the document with the id: `\"11\"`. Was expecting a `_geo.lng` field.",
+        "message": "Index `test`: Could not find longitude in the document with the id: `\"11\"`. Was expecting a `_geo.lng` field.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1833,7 +1990,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Could not find latitude in the document with the id: `\"11\"`. Was expecting a `_geo.lat` field.",
+        "message": "Index `test`: Could not find latitude in the document with the id: `\"11\"`. Was expecting a `_geo.lat` field.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1871,7 +2028,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Could not find longitude in the document with the id: `\"11\"`. Was expecting a `_geo.lng` field.",
+        "message": "Index `test`: Could not find longitude in the document with the id: `\"11\"`. Was expecting a `_geo.lng` field.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1909,7 +2066,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Could not find latitude in the document with the id: `\"11\"`. Was expecting a `_geo.lat` field.",
+        "message": "Index `test`: Could not find latitude in the document with the id: `\"11\"`. Was expecting a `_geo.lat` field.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1947,7 +2104,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Could not parse latitude nor longitude in the document with the id: `\"11\"`. Was expecting finite numbers but instead got `false` and `true`.",
+        "message": "Index `test`: Could not parse latitude nor longitude in the document with the id: `\"11\"`. Was expecting finite numbers but instead got `false` and `true`.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1985,7 +2142,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Could not find longitude in the document with the id: `\"11\"`. Was expecting a `_geo.lng` field.",
+        "message": "Index `test`: Could not find longitude in the document with the id: `\"11\"`. Was expecting a `_geo.lng` field.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -2023,7 +2180,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Could not find latitude in the document with the id: `\"11\"`. Was expecting a `_geo.lat` field.",
+        "message": "Index `test`: Could not find latitude in the document with the id: `\"11\"`. Was expecting a `_geo.lat` field.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -2061,7 +2218,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Could not parse latitude nor longitude in the document with the id: `\"11\"`. Was expecting finite numbers but instead got `\"doggo\"` and `\"doggo\"`.",
+        "message": "Index `test`: Could not parse latitude nor longitude in the document with the id: `\"11\"`. Was expecting finite numbers but instead got `\"doggo\"` and `\"doggo\"`.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -2099,7 +2256,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "The `_geo` field in the document with the id: `\"11\"` contains the following unexpected fields: `{\"doggo\":\"are the best\"}`.",
+        "message": "Index `test`: The `_geo` field in the document with the id: `\"11\"` contains the following unexpected fields: `{\"doggo\":\"are the best\"}`.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -2138,7 +2295,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Could not parse longitude in the document with the id: `\"12\"`. Was expecting a finite number but instead got `null`.",
+        "message": "Index `test`: Could not parse longitude in the document with the id: `\"12\"`. Was expecting a finite number but instead got `null`.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -2175,7 +2332,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Could not parse latitude in the document with the id: `\"12\"`. Was expecting a finite number but instead got `null`.",
+        "message": "Index `test`: Could not parse latitude in the document with the id: `\"12\"`. Was expecting a finite number but instead got `null`.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -2212,7 +2369,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Could not parse latitude nor longitude in the document with the id: `\"13\"`. Was expecting finite numbers but instead got `null` and `null`.",
+        "message": "Index `test`: Could not parse latitude nor longitude in the document with the id: `\"13\"`. Was expecting finite numbers but instead got `null` and `null`.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -2279,7 +2436,7 @@ async fn add_invalid_geo_and_then_settings() {
         ]
       },
       "error": {
-        "message": "Could not parse latitude in the document with the id: `\"11\"`. Was expecting a finite number but instead got `null`.",
+        "message": "Index `test`: Could not parse latitude in the document with the id: `\"11\"`. Was expecting a finite number but instead got `null`.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"

crates/meilisearch/tests/documents/errors.rs

@@ -604,7 +604,7 @@ async fn delete_document_by_filter() {
         "originalFilter": "\"doggo = bernese\""
       },
       "error": {
-        "message": "Attribute `doggo` is not filterable. This index does not have configured filterable attributes.\n1:6 doggo = bernese",
+        "message": "Index `EMPTY_INDEX`: Attribute `doggo` is not filterable. This index does not have configured filterable attributes.\n1:6 doggo = bernese",
         "code": "invalid_document_filter",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_filter"
@@ -636,7 +636,7 @@ async fn delete_document_by_filter() {
         "originalFilter": "\"catto = jorts\""
       },
       "error": {
-        "message": "Attribute `catto` is not filterable. Available filterable attributes are: `id`, `title`.\n1:6 catto = jorts",
+        "message": "Index `SHARED_DOCUMENTS`: Attribute `catto` is not filterable. Available filterable attributes are: `id`, `title`.\n1:6 catto = jorts",
         "code": "invalid_document_filter",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_filter"

crates/meilisearch/tests/documents/update_documents.rs

@@ -172,7 +172,7 @@ async fn error_update_documents_bad_document_id() {
     assert_eq!(
         response["error"]["message"],
         json!(
-            r#"Document identifier `"foo & bar"` is invalid. A document identifier can be of type integer or string, only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_), and can not be more than 512 bytes."#
+            r#"Document identifier `"foo & bar"` is invalid. A document identifier can be of type integer or string, only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_), and can not be more than 511 bytes."#
         )
     );
     assert_eq!(response["error"]["code"], json!("invalid_document_id"));

crates/meilisearch/tests/index/update_index.rs

@@ -95,7 +95,7 @@ async fn error_update_existing_primary_key() {
     let response = index.wait_task(2).await;
 
     let expected_response = json!({
-        "message": "Index already has a primary key: `id`.",
+        "message": "Index `test`: Index already has a primary key: `id`.",
         "code": "index_primary_key_already_exists",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#index_primary_key_already_exists"

crates/meilisearch/tests/search/errors.rs

@@ -711,7 +711,7 @@ async fn filter_invalid_attribute_array() {
     index.wait_task(task.uid()).await;
 
     let expected_response = json!({
-        "message": "Attribute `many` is not filterable. Available filterable attributes are: `title`.\n1:5 many = Glass",
+        "message": format!("Index `{}`: Attribute `many` is not filterable. Available filterable attributes are: `title`.\n1:5 many = Glass", index.uid),
         "code": "invalid_search_filter",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_search_filter"
@@ -733,7 +733,7 @@ async fn filter_invalid_attribute_string() {
     index.wait_task(task.uid()).await;
 
     let expected_response = json!({
-        "message": "Attribute `many` is not filterable. Available filterable attributes are: `title`.\n1:5 many = Glass",
+        "message": format!("Index `{}`: Attribute `many` is not filterable. Available filterable attributes are: `title`.\n1:5 many = Glass", index.uid),
         "code": "invalid_search_filter",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_search_filter"
@@ -940,7 +940,7 @@ async fn sort_unsortable_attribute() {
     index.wait_task(response.uid()).await.succeeded();
 
     let expected_response = json!({
-        "message": "Attribute `title` is not sortable. Available sortable attributes are: `id`.",
+        "message": format!("Index `{}`: Attribute `title` is not sortable. Available sortable attributes are: `id`.", index.uid),
         "code": "invalid_search_sort",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_search_sort"
@@ -998,7 +998,7 @@ async fn sort_unset_ranking_rule() {
     index.wait_task(response.uid()).await.succeeded();
 
     let expected_response = json!({
-        "message": "You must specify where `sort` is listed in the rankingRules setting to use the sort parameter at search time.",
+        "message": format!("Index `{}`: You must specify where `sort` is listed in the rankingRules setting to use the sort parameter at search time.", index.uid),
         "code": "invalid_search_sort",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_search_sort"
@@ -1024,19 +1024,18 @@ async fn search_on_unknown_field() {
         index.update_settings_searchable_attributes(json!(["id", "title"])).await;
     index.wait_task(response.uid()).await.succeeded();
 
+    let expected_response = json!({
+        "message": format!("Index `{}`: Attribute `unknown` is not searchable. Available searchable attributes are: `id, title`.", index.uid),
+        "code": "invalid_search_attributes_to_search_on",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_search_attributes_to_search_on"
+    });
     index
         .search(
             json!({"q": "Captain Marvel", "attributesToSearchOn": ["unknown"]}),
             |response, code| {
-                snapshot!(code, @"400 Bad Request");
-                snapshot!(json_string!(response), @r###"
-                {
-                  "message": "Attribute `unknown` is not searchable. Available searchable attributes are: `id, title`.",
-                  "code": "invalid_search_attributes_to_search_on",
-                  "type": "invalid_request",
-                  "link": "https://docs.meilisearch.com/errors#invalid_search_attributes_to_search_on"
-                }
-                "###);
+                assert_eq!(response, expected_response);
+                assert_eq!(code, 400);
             },
         )
         .await;
@@ -1050,19 +1049,18 @@ async fn search_on_unknown_field_plus_joker() {
         index.update_settings_searchable_attributes(json!(["id", "title"])).await;
     index.wait_task(response.uid()).await.succeeded();
 
+    let expected_response = json!({
+        "message": format!("Index `{}`: Attribute `unknown` is not searchable. Available searchable attributes are: `id, title`.", index.uid),
+        "code": "invalid_search_attributes_to_search_on",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_search_attributes_to_search_on"
+    });
     index
         .search(
             json!({"q": "Captain Marvel", "attributesToSearchOn": ["*", "unknown"]}),
             |response, code| {
-                snapshot!(code, @"400 Bad Request");
-                snapshot!(json_string!(response), @r###"
-                {
-                  "message": "Attribute `unknown` is not searchable. Available searchable attributes are: `id, title`.",
-                  "code": "invalid_search_attributes_to_search_on",
-                  "type": "invalid_request",
-                  "link": "https://docs.meilisearch.com/errors#invalid_search_attributes_to_search_on"
-                }
-                "###);
+                assert_eq!(response, expected_response);
+                assert_eq!(code, 400);
             },
         )
         .await;
@@ -1071,15 +1069,8 @@ async fn search_on_unknown_field_plus_joker() {
         .search(
             json!({"q": "Captain Marvel", "attributesToSearchOn": ["unknown", "*"]}),
             |response, code| {
-                snapshot!(code, @"400 Bad Request");
-                snapshot!(json_string!(response), @r###"
-                {
-                  "message": "Attribute `unknown` is not searchable. Available searchable attributes are: `id, title`.",
-                  "code": "invalid_search_attributes_to_search_on",
-                  "type": "invalid_request",
-                  "link": "https://docs.meilisearch.com/errors#invalid_search_attributes_to_search_on"
-                }
-                "###);
+                assert_eq!(response, expected_response);
+                assert_eq!(code, 400);
             },
         )
         .await;
@@ -1092,47 +1083,44 @@ async fn distinct_at_search_time() {
     let (task, _) = index.create(None).await;
     index.wait_task(task.uid()).await.succeeded();
 
+    let expected_response = json!({
+        "message": format!("Index `{}`: Attribute `doggo.truc` is not filterable and thus, cannot be used as distinct attribute. This index does not have configured filterable attributes.", index.uid),
+        "code": "invalid_search_distinct",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_search_distinct"
+    });
     let (response, code) =
         index.search_post(json!({"page": 0, "hitsPerPage": 2, "distinct": "doggo.truc"})).await;
-    snapshot!(code, @"400 Bad Request");
-    snapshot!(response, @r###"
-    {
-      "message": "Attribute `doggo.truc` is not filterable and thus, cannot be used as distinct attribute. This index does not have configured filterable attributes.",
-      "code": "invalid_search_distinct",
-      "type": "invalid_request",
-      "link": "https://docs.meilisearch.com/errors#invalid_search_distinct"
-    }
-    "###);
+    assert_eq!(response, expected_response);
+    assert_eq!(code, 400);
 
     let (task, _) = index.update_settings_filterable_attributes(json!(["color", "machin"])).await;
     index.wait_task(task.uid()).await;
 
+    let expected_response = json!({
+        "message": format!("Index `{}`: Attribute `doggo.truc` is not filterable and thus, cannot be used as distinct attribute. Available filterable attributes are: `color, machin`.", index.uid),
+        "code": "invalid_search_distinct",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_search_distinct"
+    });
     let (response, code) =
         index.search_post(json!({"page": 0, "hitsPerPage": 2, "distinct": "doggo.truc"})).await;
-    snapshot!(code, @"400 Bad Request");
-    snapshot!(response, @r###"
-    {
-      "message": "Attribute `doggo.truc` is not filterable and thus, cannot be used as distinct attribute. Available filterable attributes are: `color, machin`.",
-      "code": "invalid_search_distinct",
-      "type": "invalid_request",
-      "link": "https://docs.meilisearch.com/errors#invalid_search_distinct"
-    }
-    "###);
+    assert_eq!(response, expected_response);
+    assert_eq!(code, 400);
 
     let (task, _) = index.update_settings_displayed_attributes(json!(["color"])).await;
     index.wait_task(task.uid()).await;
 
+    let expected_response = json!({
+        "message": format!("Index `{}`: Attribute `doggo.truc` is not filterable and thus, cannot be used as distinct attribute. Available filterable attributes are: `color, <..hidden-attributes>`.", index.uid),
+        "code": "invalid_search_distinct",
+        "type": "invalid_request",
+        "link": "https://docs.meilisearch.com/errors#invalid_search_distinct"
+    });
     let (response, code) =
         index.search_post(json!({"page": 0, "hitsPerPage": 2, "distinct": "doggo.truc"})).await;
-    snapshot!(code, @"400 Bad Request");
-    snapshot!(response, @r###"
-    {
-      "message": "Attribute `doggo.truc` is not filterable and thus, cannot be used as distinct attribute. Available filterable attributes are: `color, <..hidden-attributes>`.",
-      "code": "invalid_search_distinct",
-      "type": "invalid_request",
-      "link": "https://docs.meilisearch.com/errors#invalid_search_distinct"
-    }
-    "###);
+    assert_eq!(response, expected_response);
+    assert_eq!(code, 400);
 
     let (response, code) =
         index.search_post(json!({"page": 0, "hitsPerPage": 2, "distinct": true})).await;

crates/meilisearch/tests/search/facet_search.rs

@@ -57,6 +57,116 @@ async fn simple_facet_search() {
     assert_eq!(response["facetHits"].as_array().unwrap().len(), 1);
 }
 
+#[actix_rt::test]
+async fn simple_facet_search_on_movies() {
+    let server = Server::new().await;
+    let index = server.index("test");
+
+    let documents = json!([
+      {
+        "id": 1,
+        "title": "Carol",
+        "genres": [
+          "Romance",
+          "Drama"
+        ],
+        "color": [
+          "red"
+        ],
+        "platforms": [
+          "MacOS",
+          "Linux",
+          "Windows"
+        ]
+      },
+      {
+        "id": 2,
+        "title": "Wonder Woman",
+        "genres": [
+          "Action",
+          "Adventure"
+        ],
+        "color": [
+          "green"
+        ],
+        "platforms": [
+          "MacOS"
+        ]
+      },
+      {
+        "id": 3,
+        "title": "Life of Pi",
+        "genres": [
+          "Adventure",
+          "Drama"
+        ],
+        "color": [
+          "blue"
+        ],
+        "platforms": [
+          "Windows"
+        ]
+      },
+      {
+        "id": 4,
+        "title": "Mad Max: Fury Road",
+        "genres": [
+          "Adventure",
+          "Science Fiction"
+        ],
+        "color": [
+          "red"
+        ],
+        "platforms": [
+          "MacOS",
+          "Linux"
+        ]
+      },
+      {
+        "id": 5,
+        "title": "Moana",
+        "genres": [
+          "Fantasy",
+          "Action"
+        ],
+        "color": [
+          "red"
+        ],
+        "platforms": [
+          "Windows"
+        ]
+      },
+      {
+        "id": 6,
+        "title": "Philadelphia",
+        "genres": [
+          "Drama"
+        ],
+        "color": [
+          "blue"
+        ],
+        "platforms": [
+          "MacOS",
+          "Linux",
+          "Windows"
+        ]
+      }
+    ]);
+    let (response, code) =
+        index.update_settings_filterable_attributes(json!(["genres", "color"])).await;
+    assert_eq!(202, code, "{:?}", response);
+    index.wait_task(response.uid()).await;
+
+    let (response, _code) = index.add_documents(documents, None).await;
+    index.wait_task(response.uid()).await;
+
+    let (response, code) =
+        index.facet_search(json!({"facetQuery": "", "facetName": "genres", "q": "" })).await;
+
+    assert_eq!(code, 200, "{}", response);
+    snapshot!(response["facetHits"], @r###"[{"value":"Action","count":2},{"value":"Adventure","count":3},{"value":"Drama","count":3},{"value":"Fantasy","count":1},{"value":"Romance","count":1},{"value":"Science Fiction","count":1}]"###);
+}
+
 #[actix_rt::test]
 async fn advanced_facet_search() {
     let server = Server::new().await;
@@ -221,8 +331,15 @@ async fn add_documents_and_deactivate_facet_search() {
     let (response, code) =
         index.facet_search(json!({"facetName": "genres", "facetQuery": "a"})).await;
 
-    assert_eq!(code, 200, "{}", response);
-    assert_eq!(dbg!(response)["facetHits"].as_array().unwrap().len(), 0);
+    assert_eq!(code, 400, "{}", response);
+    snapshot!(response, @r###"
+    {
+      "message": "The facet search is disabled for this index",
+      "code": "facet_search_disabled",
+      "type": "invalid_request",
+      "link": "https://docs.meilisearch.com/errors#facet_search_disabled"
+    }
+    "###);
 }
 
 #[actix_rt::test]
@@ -245,8 +362,15 @@ async fn deactivate_facet_search_and_add_documents() {
     let (response, code) =
         index.facet_search(json!({"facetName": "genres", "facetQuery": "a"})).await;
 
-    assert_eq!(code, 200, "{}", response);
-    assert_eq!(dbg!(response)["facetHits"].as_array().unwrap().len(), 0);
+    assert_eq!(code, 400, "{}", response);
+    snapshot!(response, @r###"
+    {
+      "message": "The facet search is disabled for this index",
+      "code": "facet_search_disabled",
+      "type": "invalid_request",
+      "link": "https://docs.meilisearch.com/errors#facet_search_disabled"
+    }
+    "###);
 }
 
 #[actix_rt::test]

crates/meilisearch/tests/search/mod.rs

@@ -1746,3 +1746,57 @@ async fn change_attributes_settings() {
         )
         .await;
 }
+
+/// Modifying facets with different casing should work correctly
+#[actix_rt::test]
+async fn change_facet_casing() {
+    let server = Server::new().await;
+    let index = server.index("test");
+
+    let (response, code) = index
+        .update_settings(json!({
+            "filterableAttributes": ["dog"],
+        }))
+        .await;
+    assert_eq!("202", code.as_str(), "{:?}", response);
+    index.wait_task(response.uid()).await;
+
+    let (response, _code) = index
+        .add_documents(
+            json!([
+                {
+                    "id": 1,
+                    "dog": "Bouvier Bernois"
+                }
+            ]),
+            None,
+        )
+        .await;
+    index.wait_task(response.uid()).await;
+
+    let (response, _code) = index
+        .add_documents(
+            json!([
+                {
+                    "id": 1,
+                    "dog": "bouvier bernois"
+                }
+            ]),
+            None,
+        )
+        .await;
+    index.wait_task(response.uid()).await;
+
+    index
+        .search(json!({ "facets": ["dog"] }), |response, code| {
+            meili_snap::snapshot!(code, @"200 OK");
+            meili_snap::snapshot!(meili_snap::json_string!(response["facetDistribution"]), @r###"
+            {
+              "dog": {
+                "bouvier bernois": 1
+              }
+            }
+            "###);
+        })
+        .await;
+}

crates/meilisearch/tests/search/multi.rs

@@ -1070,7 +1070,7 @@ async fn federation_one_query_error() {
     snapshot!(code, @"400 Bad Request");
     snapshot!(json_string!(response), @r###"
     {
-      "message": "Inside `.queries[1]`: Attribute `title` is not filterable. This index does not have configured filterable attributes.\n1:6 title = toto",
+      "message": "Inside `.queries[1]`: Index `nested`: Attribute `title` is not filterable. This index does not have configured filterable attributes.\n1:6 title = toto",
       "code": "invalid_search_filter",
       "type": "invalid_request",
       "link": "https://docs.meilisearch.com/errors#invalid_search_filter"
@@ -1102,7 +1102,7 @@ async fn federation_one_query_sort_error() {
     snapshot!(code, @"400 Bad Request");
     snapshot!(json_string!(response), @r###"
     {
-      "message": "Inside `.queries[1]`: Attribute `doggos` is not sortable. This index does not have configured sortable attributes.",
+      "message": "Inside `.queries[1]`: Index `nested`: Attribute `doggos` is not sortable. This index does not have configured sortable attributes.",
       "code": "invalid_search_sort",
       "type": "invalid_request",
       "link": "https://docs.meilisearch.com/errors#invalid_search_sort"
@@ -1166,7 +1166,7 @@ async fn federation_multiple_query_errors() {
     snapshot!(code, @"400 Bad Request");
     snapshot!(json_string!(response), @r###"
     {
-      "message": "Inside `.queries[0]`: Attribute `title` is not filterable. This index does not have configured filterable attributes.\n1:6 title = toto",
+      "message": "Inside `.queries[0]`: Index `test`: Attribute `title` is not filterable. This index does not have configured filterable attributes.\n1:6 title = toto",
       "code": "invalid_search_filter",
       "type": "invalid_request",
       "link": "https://docs.meilisearch.com/errors#invalid_search_filter"
@@ -1198,7 +1198,7 @@ async fn federation_multiple_query_sort_errors() {
     snapshot!(code, @"400 Bad Request");
     snapshot!(json_string!(response), @r###"
     {
-      "message": "Inside `.queries[0]`: Attribute `title` is not sortable. This index does not have configured sortable attributes.",
+      "message": "Inside `.queries[0]`: Index `test`: Attribute `title` is not sortable. This index does not have configured sortable attributes.",
       "code": "invalid_search_sort",
       "type": "invalid_request",
       "link": "https://docs.meilisearch.com/errors#invalid_search_sort"
@@ -1231,7 +1231,7 @@ async fn federation_multiple_query_errors_interleaved() {
     snapshot!(code, @"400 Bad Request");
     snapshot!(json_string!(response), @r###"
     {
-      "message": "Inside `.queries[1]`: Attribute `doggos` is not filterable. This index does not have configured filterable attributes.\n1:7 doggos IN [intel, kefir]",
+      "message": "Inside `.queries[1]`: Index `nested`: Attribute `doggos` is not filterable. This index does not have configured filterable attributes.\n1:7 doggos IN [intel, kefir]",
       "code": "invalid_search_filter",
       "type": "invalid_request",
       "link": "https://docs.meilisearch.com/errors#invalid_search_filter"
@@ -1264,7 +1264,7 @@ async fn federation_multiple_query_sort_errors_interleaved() {
     snapshot!(code, @"400 Bad Request");
     snapshot!(json_string!(response), @r###"
     {
-      "message": "Inside `.queries[1]`: Attribute `doggos` is not sortable. This index does not have configured sortable attributes.",
+      "message": "Inside `.queries[1]`: Index `nested`: Attribute `doggos` is not sortable. This index does not have configured sortable attributes.",
       "code": "invalid_search_sort",
       "type": "invalid_request",
       "link": "https://docs.meilisearch.com/errors#invalid_search_sort"

crates/meilisearch/tests/settings/get_settings.rs

@@ -1,44 +1,185 @@
-use std::collections::HashMap;
-
-use once_cell::sync::Lazy;
-
-use crate::common::{Server, Value};
+use crate::common::Server;
 use crate::json;
 
-static DEFAULT_SETTINGS_VALUES: Lazy<HashMap<&'static str, Value>> = Lazy::new(|| {
-    let mut map = HashMap::new();
-    map.insert("displayed_attributes", json!(["*"]));
-    map.insert("searchable_attributes", json!(["*"]));
-    map.insert("localized_attributes", json!(null));
-    map.insert("filterable_attributes", json!([]));
-    map.insert("distinct_attribute", json!(null));
-    map.insert(
-        "ranking_rules",
-        json!(["words", "typo", "proximity", "attribute", "sort", "exactness"]),
-    );
-    map.insert("stop_words", json!([]));
-    map.insert("non_separator_tokens", json!([]));
-    map.insert("separator_tokens", json!([]));
-    map.insert("dictionary", json!([]));
-    map.insert("synonyms", json!({}));
-    map.insert(
-        "faceting",
-        json!({
-            "maxValuesPerFacet": json!(100),
-            "sortFacetValuesBy": {
-                "*": "alpha"
+macro_rules! test_setting_routes {
+    ($({setting: $setting:ident, update_verb: $update_verb:ident, default_value: $default_value:tt},) *) => {
+        $(
+            mod $setting {
+                use crate::common::Server;
+
+                #[actix_rt::test]
+                async fn get_unexisting_index() {
+                    let server = Server::new().await;
+                    let url = format!("/indexes/test/settings/{}",
+                        stringify!($setting)
+                        .chars()
+                        .map(|c| if c == '_' { '-' } else { c })
+                        .collect::<String>());
+                    let (_response, code) = server.service.get(url).await;
+                    assert_eq!(code, 404);
+                }
+
+                #[actix_rt::test]
+                async fn update_unexisting_index() {
+                    let server = Server::new().await;
+                    let url = format!("/indexes/test/settings/{}",
+                        stringify!($setting)
+                        .chars()
+                        .map(|c| if c == '_' { '-' } else { c })
+                        .collect::<String>());
+                    let (response, code) = server.service.$update_verb(url, serde_json::Value::Null.into()).await;
+                    assert_eq!(code, 202, "{}", response);
+                    server.index("").wait_task(0).await;
+                    let (response, code) = server.index("test").get().await;
+                    assert_eq!(code, 200, "{}", response);
+                }
+
+                #[actix_rt::test]
+                async fn delete_unexisting_index() {
+                    let server = Server::new().await;
+                    let url = format!("/indexes/test/settings/{}",
+                        stringify!($setting)
+                        .chars()
+                        .map(|c| if c == '_' { '-' } else { c })
+                        .collect::<String>());
+                    let (_, code) = server.service.delete(url).await;
+                    assert_eq!(code, 202);
+                    let response = server.index("").wait_task(0).await;
+                    assert_eq!(response["status"], "failed");
+                }
+
+                #[actix_rt::test]
+                async fn get_default() {
+                    let server = Server::new().await;
+                    let index = server.index("test");
+                    let (response, code) = index.create(None).await;
+                    assert_eq!(code, 202, "{}", response);
+                    index.wait_task(0).await;
+                    let url = format!("/indexes/test/settings/{}",
+                        stringify!($setting)
+                        .chars()
+                        .map(|c| if c == '_' { '-' } else { c })
+                        .collect::<String>());
+                    let (response, code) = server.service.get(url).await;
+                    assert_eq!(code, 200, "{}", response);
+                    let expected = crate::json!($default_value);
+                    assert_eq!(expected, response);
+                }
             }
-        }),
-    );
-    map.insert(
-        "pagination",
-        json!({
-            "maxTotalHits": json!(1000),
-        }),
-    );
-    map.insert("search_cutoff_ms", json!(null));
-    map
-});
+        )*
+
+        #[actix_rt::test]
+        async fn all_setting_tested() {
+            let expected = std::collections::BTreeSet::from_iter(meilisearch::routes::indexes::settings::ALL_SETTINGS_NAMES.iter());
+            let tested = std::collections::BTreeSet::from_iter([$(stringify!($setting)),*].iter());
+            let diff: Vec<_> = expected.difference(&tested).collect();
+            assert!(diff.is_empty(), "Not all settings were tested, please add the following settings to the `test_setting_routes!` macro: {:?}", diff);
+        }
+    };
+}
+
+test_setting_routes!(
+    {
+        setting: filterable_attributes,
+        update_verb: put,
+        default_value: []
+    },
+    {
+        setting: displayed_attributes,
+        update_verb: put,
+        default_value: ["*"]
+    },
+    {
+        setting: localized_attributes,
+        update_verb: put,
+        default_value: null
+    },
+    {
+        setting: searchable_attributes,
+        update_verb: put,
+        default_value: ["*"]
+    },
+    {
+        setting: distinct_attribute,
+        update_verb: put,
+        default_value: null
+    },
+    {
+        setting: stop_words,
+        update_verb: put,
+        default_value: []
+    },
+    {
+        setting: separator_tokens,
+        update_verb: put,
+        default_value: []
+    },
+    {
+        setting: non_separator_tokens,
+        update_verb: put,
+        default_value: []
+    },
+    {
+        setting: dictionary,
+        update_verb: put,
+        default_value: []
+    },
+    {
+        setting: ranking_rules,
+        update_verb: put,
+        default_value: ["words", "typo", "proximity", "attribute", "sort", "exactness"]
+    },
+    {
+        setting: synonyms,
+        update_verb: put,
+        default_value: {}
+    },
+    {
+        setting: pagination,
+        update_verb: patch,
+        default_value: {"maxTotalHits": 1000}
+    },
+    {
+        setting: faceting,
+        update_verb: patch,
+        default_value: {"maxValuesPerFacet": 100, "sortFacetValuesBy": {"*": "alpha"}}
+    },
+    {
+        setting: search_cutoff_ms,
+        update_verb: put,
+        default_value: null
+    },
+    {
+        setting: embedders,
+        update_verb: patch,
+        default_value: null
+    },
+    {
+        setting: facet_search,
+        update_verb: put,
+        default_value: true
+    },
+    {
+        setting: prefix_search,
+        update_verb: put,
+        default_value: "indexingTime"
+    },
+    {
+        setting: proximity_precision,
+        update_verb: put,
+        default_value: "byWord"
+    },
+    {
+        setting: sortable_attributes,
+        update_verb: put,
+        default_value: []
+    },
+    {
+        setting: typo_tolerance,
+        update_verb: patch,
+        default_value: {"enabled": true, "minWordSizeForTypos": {"oneTypo": 5, "twoTypos": 9}, "disableOnWords": [], "disableOnAttributes": []}
+    },
+);
 
 #[actix_rt::test]
 async fn get_settings_unexisting_index() {
@@ -342,93 +483,6 @@ async fn error_update_setting_unexisting_index_invalid_uid() {
     "###);
 }
 
-macro_rules! test_setting_routes {
-    ($($setting:ident $write_method:ident), *) => {
-        $(
-            mod $setting {
-                use crate::common::Server;
-                use super::DEFAULT_SETTINGS_VALUES;
-
-                #[actix_rt::test]
-                async fn get_unexisting_index() {
-                    let server = Server::new().await;
-                    let url = format!("/indexes/test/settings/{}",
-                        stringify!($setting)
-                        .chars()
-                        .map(|c| if c == '_' { '-' } else { c })
-                        .collect::<String>());
-                    let (_response, code) = server.service.get(url).await;
-                    assert_eq!(code, 404);
-                }
-
-                #[actix_rt::test]
-                async fn update_unexisting_index() {
-                    let server = Server::new().await;
-                    let url = format!("/indexes/test/settings/{}",
-                        stringify!($setting)
-                        .chars()
-                        .map(|c| if c == '_' { '-' } else { c })
-                        .collect::<String>());
-                    let (response, code) = server.service.$write_method(url, serde_json::Value::Null.into()).await;
-                    assert_eq!(code, 202, "{}", response);
-                    server.index("").wait_task(0).await;
-                    let (response, code) = server.index("test").get().await;
-                    assert_eq!(code, 200, "{}", response);
-                }
-
-                #[actix_rt::test]
-                async fn delete_unexisting_index() {
-                    let server = Server::new().await;
-                    let url = format!("/indexes/test/settings/{}",
-                        stringify!($setting)
-                        .chars()
-                        .map(|c| if c == '_' { '-' } else { c })
-                        .collect::<String>());
-                    let (_, code) = server.service.delete(url).await;
-                    assert_eq!(code, 202);
-                    let response = server.index("").wait_task(0).await;
-                    assert_eq!(response["status"], "failed");
-                }
-
-                #[actix_rt::test]
-                async fn get_default() {
-                    let server = Server::new().await;
-                    let index = server.index("test");
-                    let (response, code) = index.create(None).await;
-                    assert_eq!(code, 202, "{}", response);
-                    index.wait_task(0).await;
-                    let url = format!("/indexes/test/settings/{}",
-                        stringify!($setting)
-                        .chars()
-                        .map(|c| if c == '_' { '-' } else { c })
-                        .collect::<String>());
-                    let (response, code) = server.service.get(url).await;
-                    assert_eq!(code, 200, "{}", response);
-                    let expected = DEFAULT_SETTINGS_VALUES.get(stringify!($setting)).unwrap();
-                    assert_eq!(expected, &response);
-                }
-            }
-        )*
-    };
-}
-
-test_setting_routes!(
-    filterable_attributes put,
-    displayed_attributes put,
-    localized_attributes put,
-    searchable_attributes put,
-    distinct_attribute put,
-    stop_words put,
-    separator_tokens put,
-    non_separator_tokens put,
-    dictionary put,
-    ranking_rules put,
-    synonyms put,
-    pagination patch,
-    faceting patch,
-    search_cutoff_ms put
-);
-
 #[actix_rt::test]
 async fn error_set_invalid_ranking_rules() {
     let server = Server::new().await;

crates/meilisearch/tests/similar/errors.rs

@@ -79,7 +79,7 @@ async fn similar_bad_id() {
     snapshot!(code, @"400 Bad Request");
     snapshot!(json_string!(response), @r###"
     {
-      "message": "Invalid value at `.id`: the value of `id` is invalid. A document identifier can be of type integer or string, only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_), and can not be more than 512 bytes.",
+      "message": "Invalid value at `.id`: the value of `id` is invalid. A document identifier can be of type integer or string, only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_), and can not be more than 511 bytes.",
       "code": "invalid_similar_id",
       "type": "invalid_request",
       "link": "https://docs.meilisearch.com/errors#invalid_similar_id"
@@ -172,7 +172,7 @@ async fn similar_invalid_id() {
     snapshot!(code, @"400 Bad Request");
     snapshot!(json_string!(response), @r###"
     {
-      "message": "Invalid value at `.id`: the value of `id` is invalid. A document identifier can be of type integer or string, only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_), and can not be more than 512 bytes.",
+      "message": "Invalid value at `.id`: the value of `id` is invalid. A document identifier can be of type integer or string, only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_), and can not be more than 511 bytes.",
       "code": "invalid_similar_id",
       "type": "invalid_request",
       "link": "https://docs.meilisearch.com/errors#invalid_similar_id"

crates/meilisearch/tests/snapshot/mod.rs

@@ -129,11 +129,11 @@ async fn perform_on_demand_snapshot() {
 
     index.load_test_set().await;
 
-    server.index("doggo").create(Some("bone")).await;
-    index.wait_task(2).await;
+    let (task, _) = server.index("doggo").create(Some("bone")).await;
+    index.wait_task(task.uid()).await.succeeded();
 
-    server.index("doggo").create(Some("bone")).await;
-    index.wait_task(2).await;
+    let (task, _) = server.index("doggo").create(Some("bone")).await;
+    index.wait_task(task.uid()).await.failed();
 
     let (task, code) = server.create_snapshot().await;
     snapshot!(code, @"202 Accepted");

crates/meilisearch/tests/tasks/mod.rs

@@ -448,7 +448,7 @@ async fn test_summarized_delete_documents_by_filter() {
         "originalFilter": "\"doggo = bernese\""
       },
       "error": {
-        "message": "Attribute `doggo` is not filterable. This index does not have configured filterable attributes.\n1:6 doggo = bernese",
+        "message": "Index `test`: Attribute `doggo` is not filterable. This index does not have configured filterable attributes.\n1:6 doggo = bernese",
         "code": "invalid_document_filter",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_filter"

crates/meilisearch/tests/vector/binary_quantized.rs

@@ -318,7 +318,7 @@ async fn try_to_disable_binary_quantization() {
         }
       },
       "error": {
-        "message": "`.embedders.manual.binaryQuantized`: Cannot disable the binary quantization.\n - Note: Binary quantization is a lossy operation that cannot be reverted.\n - Hint: Add a new embedder that is non-quantized and regenerate the vectors.",
+        "message": "Index `doggo`: `.embedders.manual.binaryQuantized`: Cannot disable the binary quantization.\n - Note: Binary quantization is a lossy operation that cannot be reverted.\n - Hint: Add a new embedder that is non-quantized and regenerate the vectors.",
         "code": "invalid_settings_embedders",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_settings_embedders"

crates/meilisearch/tests/vector/mod.rs

@@ -250,7 +250,7 @@ async fn user_provided_embeddings_error() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Bad embedder configuration in the document with id: `0`. Missing field `._vectors.manual.regenerate`\n  - note: `._vectors.manual` must be an array of floats, an array of arrays of floats, or an object with field `regenerate`",
+        "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Missing field `._vectors.manual.regenerate`\n  - note: `._vectors.manual` must be an array of floats, an array of arrays of floats, or an object with field `regenerate`",
         "code": "invalid_vectors_type",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -280,7 +280,7 @@ async fn user_provided_embeddings_error() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Bad embedder configuration in the document with id: `0`. Missing field `._vectors.manual.regenerate`\n  - note: `._vectors.manual` must be an array of floats, an array of arrays of floats, or an object with field `regenerate`",
+        "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Missing field `._vectors.manual.regenerate`\n  - note: `._vectors.manual` must be an array of floats, an array of arrays of floats, or an object with field `regenerate`",
         "code": "invalid_vectors_type",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -311,7 +311,7 @@ async fn user_provided_embeddings_error() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Bad embedder configuration in the document with id: `0`. Could not parse `._vectors.manual.regenerate`: invalid type: string \"yes please\", expected a boolean at line 1 column 26",
+        "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Could not parse `._vectors.manual.regenerate`: invalid type: string \"yes please\", expected a boolean at line 1 column 26",
         "code": "invalid_vectors_type",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -340,7 +340,7 @@ async fn user_provided_embeddings_error() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings`: expected null or an array, but found a boolean: `true`",
+        "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings`: expected null or an array, but found a boolean: `true`",
         "code": "invalid_vectors_type",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -369,7 +369,7 @@ async fn user_provided_embeddings_error() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[0]`: expected a number or an array, but found a boolean: `true`",
+        "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[0]`: expected a number or an array, but found a boolean: `true`",
         "code": "invalid_vectors_type",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -398,7 +398,7 @@ async fn user_provided_embeddings_error() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[0][0]`: expected a number, but found a boolean: `true`",
+        "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[0][0]`: expected a number, but found a boolean: `true`",
         "code": "invalid_vectors_type",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -440,7 +440,7 @@ async fn user_provided_embeddings_error() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[1]`: expected a number, but found an array: `[0.2,0.3]`",
+        "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[1]`: expected a number, but found an array: `[0.2,0.3]`",
         "code": "invalid_vectors_type",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -469,7 +469,7 @@ async fn user_provided_embeddings_error() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[1]`: expected an array, but found a number: `0.3`",
+        "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[1]`: expected an array, but found a number: `0.3`",
         "code": "invalid_vectors_type",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -498,7 +498,7 @@ async fn user_provided_embeddings_error() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[0][1]`: expected a number, but found a boolean: `true`",
+        "message": "Index `doggo`: Bad embedder configuration in the document with id: `0`. Invalid value type at `._vectors.manual.embeddings[0][1]`: expected a number, but found a boolean: `true`",
         "code": "invalid_vectors_type",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_vectors_type"
@@ -539,7 +539,7 @@ async fn user_provided_vectors_error() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "While embedding documents for embedder `manual`: no vectors provided for document `40` and at least 4 other document(s)\n- Note: `manual` has `source: userProvided`, so documents must provide embeddings as an array in `_vectors.manual`.\n- Hint: opt-out for a document with `_vectors.manual: null`",
+        "message": "Index `doggo`: While embedding documents for embedder `manual`: no vectors provided for document `40` and at least 4 other document(s)\n- Note: `manual` has `source: userProvided`, so documents must provide embeddings as an array in `_vectors.manual`.\n- Hint: opt-out for a document with `_vectors.manual: null`",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -569,7 +569,7 @@ async fn user_provided_vectors_error() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "While embedding documents for embedder `manual`: no vectors provided for document `42`\n- Note: `manual` has `source: userProvided`, so documents must provide embeddings as an array in `_vectors.manual`.\n- Hint: try replacing `_vector` by `_vectors` in 1 document(s).",
+        "message": "Index `doggo`: While embedding documents for embedder `manual`: no vectors provided for document `42`\n- Note: `manual` has `source: userProvided`, so documents must provide embeddings as an array in `_vectors.manual`.\n- Hint: try replacing `_vector` by `_vectors` in 1 document(s).",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -599,7 +599,7 @@ async fn user_provided_vectors_error() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "While embedding documents for embedder `manual`: no vectors provided for document `42`\n- Note: `manual` has `source: userProvided`, so documents must provide embeddings as an array in `_vectors.manual`.\n- Hint: try replacing `_vectors.manaul` by `_vectors.manual` in 1 document(s).",
+        "message": "Index `doggo`: While embedding documents for embedder `manual`: no vectors provided for document `42`\n- Note: `manual` has `source: userProvided`, so documents must provide embeddings as an array in `_vectors.manual`.\n- Hint: try replacing `_vectors.manaul` by `_vectors.manual` in 1 document(s).",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"

crates/meilisearch/tests/vector/openai.rs

@@ -713,7 +713,7 @@ async fn bad_api_key() {
         }
       },
       "error": {
-        "message": "While embedding documents for embedder `default`: user error: could not authenticate against OpenAI server\n  - server replied with `{\"error\":{\"message\":\"Incorrect API key provided: Bearer doggo. You can find your API key at https://platform.openai.com/account/api-keys.\",\"type\":\"invalid_request_error\",\"param\":null,\"code\":\"invalid_api_key\"}}`\n  - Hint: Check the `apiKey` parameter in the embedder configuration, and the `MEILI_OPENAI_API_KEY` and `OPENAI_API_KEY` environment variables",
+        "message": "Index `doggo`: While embedding documents for embedder `default`: user error: could not authenticate against OpenAI server\n  - server replied with `{\"error\":{\"message\":\"Incorrect API key provided: Bearer doggo. You can find your API key at https://platform.openai.com/account/api-keys.\",\"type\":\"invalid_request_error\",\"param\":null,\"code\":\"invalid_api_key\"}}`\n  - Hint: Check the `apiKey` parameter in the embedder configuration, and the `MEILI_OPENAI_API_KEY` and `OPENAI_API_KEY` environment variables",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -757,7 +757,7 @@ async fn bad_api_key() {
         }
       },
       "error": {
-        "message": "While embedding documents for embedder `default`: user error: could not authenticate against OpenAI server\n  - server replied with `{\"error\":{\"message\":\"You didn't provide an API key. You need to provide your API key in an Authorization header using Bearer auth (i.e. Authorization: Bearer YOUR_KEY), or as the password field (with blank username) if you're accessing the API from your browser and are prompted for a username and password. You can obtain an API key from https://platform.openai.com/account/api-keys.\",\"type\":\"invalid_request_error\",\"param\":null,\"code\":null}}`\n  - Hint: Check the `apiKey` parameter in the embedder configuration, and the `MEILI_OPENAI_API_KEY` and `OPENAI_API_KEY` environment variables",
+        "message": "Index `doggo`: While embedding documents for embedder `default`: user error: could not authenticate against OpenAI server\n  - server replied with `{\"error\":{\"message\":\"You didn't provide an API key. You need to provide your API key in an Authorization header using Bearer auth (i.e. Authorization: Bearer YOUR_KEY), or as the password field (with blank username) if you're accessing the API from your browser and are prompted for a username and password. You can obtain an API key from https://platform.openai.com/account/api-keys.\",\"type\":\"invalid_request_error\",\"param\":null,\"code\":null}}`\n  - Hint: Check the `apiKey` parameter in the embedder configuration, and the `MEILI_OPENAI_API_KEY` and `OPENAI_API_KEY` environment variables",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"

crates/meilisearch/tests/vector/rest.rs

@@ -985,7 +985,7 @@ async fn bad_settings() {
         }
       },
       "error": {
-        "message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with runtime error: error extracting embeddings from the response:\n  - in `response`, while extracting a single \"{{embedding}}\", expected `response` to be an array of numbers, but failed to parse server response:\n  - invalid type: map, expected a sequence",
+        "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with runtime error: error extracting embeddings from the response:\n  - in `response`, while extracting a single \"{{embedding}}\", expected `response` to be an array of numbers, but failed to parse server response:\n  - invalid type: map, expected a sequence",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1025,7 +1025,7 @@ async fn bad_settings() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "While embedding documents for embedder `rest`: runtime error: was expecting embeddings of dimension `2`, got embeddings of dimensions `3`",
+        "message": "Index `doggo`: While embedding documents for embedder `rest`: runtime error: was expecting embeddings of dimension `2`, got embeddings of dimensions `3`",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1178,7 +1178,7 @@ async fn server_returns_bad_request() {
         }
       },
       "error": {
-        "message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with user error: sent a bad request to embedding server\n  - Hint: check that the `request` in the embedder configuration matches the remote server's API\n  - server replied with `{\"error\":\"Invalid request: invalid type: string \\\"test\\\", expected struct MultipleRequest at line 1 column 6\"}`",
+        "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with user error: sent a bad request to embedding server\n  - Hint: check that the `request` in the embedder configuration matches the remote server's API\n  - server replied with `{\"error\":\"Invalid request: invalid type: string \\\"test\\\", expected struct MultipleRequest at line 1 column 6\"}`",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1247,7 +1247,7 @@ async fn server_returns_bad_request() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "While embedding documents for embedder `rest`: user error: sent a bad request to embedding server\n  - Hint: check that the `request` in the embedder configuration matches the remote server's API\n  - server replied with `{\"error\":\"Invalid request: invalid type: string \\\"name: kefir\\\\n\\\", expected struct MultipleRequest at line 1 column 15\"}`",
+        "message": "Index `doggo`: While embedding documents for embedder `rest`: user error: sent a bad request to embedding server\n  - Hint: check that the `request` in the embedder configuration matches the remote server's API\n  - server replied with `{\"error\":\"Invalid request: invalid type: string \\\"name: kefir\\\\n\\\", expected struct MultipleRequest at line 1 column 15\"}`",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1306,7 +1306,7 @@ async fn server_returns_bad_response() {
         }
       },
       "error": {
-        "message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with runtime error: error extracting embeddings from the response:\n  - in `response`, while extracting the array of \"{{embedding}}\"s, configuration expects `response` to be an array with at least 1 item(s) but server sent an object with 1 field(s)",
+        "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with runtime error: error extracting embeddings from the response:\n  - in `response`, while extracting the array of \"{{embedding}}\"s, configuration expects `response` to be an array with at least 1 item(s) but server sent an object with 1 field(s)",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1362,7 +1362,7 @@ async fn server_returns_bad_response() {
         }
       },
       "error": {
-        "message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with runtime error: error extracting embeddings from the response:\n  - in `response`, while extracting item #0 from the array of \"{{embedding}}\"s, expected `response` to be an array of numbers, but failed to parse server response:\n  - invalid type: map, expected a sequence",
+        "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with runtime error: error extracting embeddings from the response:\n  - in `response`, while extracting item #0 from the array of \"{{embedding}}\"s, expected `response` to be an array of numbers, but failed to parse server response:\n  - invalid type: map, expected a sequence",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1414,7 +1414,7 @@ async fn server_returns_bad_response() {
         }
       },
       "error": {
-        "message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with runtime error: error extracting embeddings from the response:\n  - in `response.output`, while extracting a single \"{{embedding}}\", expected `output` to be an array of numbers, but failed to parse server response:\n  - invalid type: map, expected f32",
+        "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with runtime error: error extracting embeddings from the response:\n  - in `response.output`, while extracting a single \"{{embedding}}\", expected `output` to be an array of numbers, but failed to parse server response:\n  - invalid type: map, expected f32",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1478,7 +1478,7 @@ async fn server_returns_bad_response() {
         }
       },
       "error": {
-        "message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with runtime error: error extracting embeddings from the response:\n  - in `response.embedding`, while extracting item #0 from the array of \"{{embedding}}\"s, configuration expects `embedding` to be an object with key `data` but server sent an array of size 3",
+        "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with runtime error: error extracting embeddings from the response:\n  - in `response.embedding`, while extracting item #0 from the array of \"{{embedding}}\"s, configuration expects `embedding` to be an object with key `data` but server sent an array of size 3",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1542,7 +1542,7 @@ async fn server_returns_bad_response() {
         }
       },
       "error": {
-        "message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with runtime error: error extracting embeddings from the response:\n  - in `response.output[0]`, while extracting a single \"{{embedding}}\", configuration expects key \"embeddings\", which is missing in response\n  - Hint: item #0 inside `output` has key `embedding`, did you mean `response.output[0].embedding` in embedder configuration?",
+        "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with runtime error: error extracting embeddings from the response:\n  - in `response.output[0]`, while extracting a single \"{{embedding}}\", configuration expects key \"embeddings\", which is missing in response\n  - Hint: item #0 inside `output` has key `embedding`, did you mean `response.output[0].embedding` in embedder configuration?",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1908,7 +1908,7 @@ async fn server_custom_header() {
         }
       },
       "error": {
-        "message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with user error: could not authenticate against embedding server\n  - server replied with `{\"error\":\"missing header 'my-nonstandard-auth'\"}`\n  - Hint: Check the `apiKey` parameter in the embedder configuration",
+        "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with user error: could not authenticate against embedding server\n  - server replied with `{\"error\":\"missing header 'my-nonstandard-auth'\"}`\n  - Hint: Check the `apiKey` parameter in the embedder configuration",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -1951,7 +1951,7 @@ async fn server_custom_header() {
         }
       },
       "error": {
-        "message": "Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with user error: could not authenticate against embedding server\n  - server replied with `{\"error\":\"thou shall not pass, Balrog\"}`\n  - Hint: Check the `apiKey` parameter in the embedder configuration",
+        "message": "Index `doggo`: Error while generating embeddings: runtime error: could not determine model dimensions:\n  - test embedding failed with user error: could not authenticate against embedding server\n  - server replied with `{\"error\":\"thou shall not pass, Balrog\"}`\n  - Hint: Check the `apiKey` parameter in the embedder configuration",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"
@@ -2099,7 +2099,7 @@ async fn searchable_reindex() {
         ]
       },
       "error": {
-        "message": "While embedding documents for embedder `rest`: error: received unexpected HTTP 404 from embedding server\n  - server replied with `{\"error\":\"text not found\",\"text\":\"breed: patou\\n\"}`",
+        "message": "Index `doggo`: While embedding documents for embedder `rest`: error: received unexpected HTTP 404 from embedding server\n  - server replied with `{\"error\":\"text not found\",\"text\":\"breed: patou\\n\"}`",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"

crates/meilitool/Cargo.toml

@@ -10,12 +10,15 @@ license.workspace = true
 
 [dependencies]
 anyhow = "1.0.86"
+arroy_v04_to_v05 = { package = "arroy", git = "https://github.com/meilisearch/arroy/", tag = "DO-NOT-DELETE-upgrade-v04-to-v05" }
 clap = { version = "4.5.9", features = ["derive"] }
 dump = { path = "../dump" }
 file-store = { path = "../file-store" }
+indexmap = { version = "2.7.0", features = ["serde"] }
 meilisearch-auth = { path = "../meilisearch-auth" }
 meilisearch-types = { path = "../meilisearch-types" }
 serde = { version = "1.0.209", features = ["derive"] }
+serde_json = { version = "1.0.133", features = ["preserve_order"] }
+tempfile = "3.14.0"
 time = { version = "0.3.36", features = ["formatting", "parsing", "alloc"] }
 uuid = { version = "1.10.0", features = ["v4"], default-features = false }
-arroy_v04_to_v05 = { package = "arroy", git = "https://github.com/meilisearch/arroy/", tag = "DO-NOT-DELETE-upgrade-v04-to-v05" }

crates/meilitool/src/main.rs

@@ -73,7 +73,7 @@ enum Command {
     ///
     /// Supported upgrade paths:
     ///
-    /// - v1.9.x -> v1.10.x -> v1.11.x
+    /// - v1.9.x -> v1.10.x -> v1.11.x -> v1.12.x
     OfflineUpgrade {
         #[arg(long)]
         target_version: String,
@@ -88,7 +88,7 @@ fn main() -> anyhow::Result<()> {
     match command {
         Command::ClearTaskQueue => clear_task_queue(db_path),
         Command::ExportADump { dump_dir, skip_enqueued_tasks } => {
-            export_a_dump(db_path, dump_dir, skip_enqueued_tasks)
+            export_a_dump(db_path, dump_dir, skip_enqueued_tasks, detected_version)
         }
         Command::OfflineUpgrade { target_version } => {
             let target_version = parse_version(&target_version).context("While parsing `--target-version`. Make sure `--target-version` is in the format MAJOR.MINOR.PATCH")?;
@@ -187,6 +187,7 @@ fn export_a_dump(
     db_path: PathBuf,
     dump_dir: PathBuf,
     skip_enqueued_tasks: bool,
+    detected_version: (String, String, String),
 ) -> Result<(), anyhow::Error> {
     let started_at = OffsetDateTime::now_utc();
 
@@ -238,9 +239,6 @@ fn export_a_dump(
     if skip_enqueued_tasks {
         eprintln!("Skip dumping the enqueued tasks...");
     } else {
-        eprintln!("Dumping the enqueued tasks...");
-
-        // 3. dump the tasks
         let mut dump_tasks = dump.create_tasks_queue()?;
         let mut count = 0;
         for ret in all_tasks.iter(&rtxn)? {
@@ -254,18 +252,39 @@ fn export_a_dump(
                 if status == Status::Enqueued {
                     let content_file = file_store.get_update(content_file_uuid)?;
 
-                    let reader =
-                        DocumentsBatchReader::from_reader(content_file).with_context(|| {
-                            format!("While reading content file {:?}", content_file_uuid)
-                        })?;
-
-                    let (mut cursor, documents_batch_index) = reader.into_cursor_and_fields_index();
-                    while let Some(doc) = cursor.next_document().with_context(|| {
-                        format!("While iterating on content file {:?}", content_file_uuid)
-                    })? {
-                        dump_content_file
-                            .push_document(&obkv_to_object(doc, &documents_batch_index)?)?;
+                    if (
+                        detected_version.0.as_str(),
+                        detected_version.1.as_str(),
+                        detected_version.2.as_str(),
+                    ) < ("1", "12", "0")
+                    {
+                        eprintln!("Dumping the enqueued tasks reading them in obkv format...");
+                        let reader =
+                            DocumentsBatchReader::from_reader(content_file).with_context(|| {
+                                format!("While reading content file {:?}", content_file_uuid)
+                            })?;
+                        let (mut cursor, documents_batch_index) =
+                            reader.into_cursor_and_fields_index();
+                        while let Some(doc) = cursor.next_document().with_context(|| {
+                            format!("While iterating on content file {:?}", content_file_uuid)
+                        })? {
+                            dump_content_file
+                                .push_document(&obkv_to_object(doc, &documents_batch_index)?)?;
+                        }
+                    } else {
+                        eprintln!(
+                            "Dumping the enqueued tasks reading them in JSON stream format..."
+                        );
+                        for document in
+                            serde_json::de::Deserializer::from_reader(content_file).into_iter()
+                        {
+                            let document = document.with_context(|| {
+                                format!("While reading content file {:?}", content_file_uuid)
+                            })?;
+                            dump_content_file.push_document(&document)?;
+                        }
                     }
+
                     dump_content_file.flush()?;
                     count += 1;
                 }

crates/meilitool/src/upgrade/mod.rs

@@ -1,13 +1,14 @@
 mod v1_10;
 mod v1_11;
+mod v1_12;
 mod v1_9;
 
 use std::path::{Path, PathBuf};
 
 use anyhow::{bail, Context};
 use meilisearch_types::versioning::create_version_file;
-
 use v1_10::v1_9_to_v1_10;
+use v1_12::{v1_11_to_v1_12, v1_12_to_v1_12_3};
 
 use crate::upgrade::v1_11::v1_10_to_v1_11;
 
@@ -19,11 +20,48 @@ pub struct OfflineUpgrade {
 
 impl OfflineUpgrade {
     pub fn upgrade(self) -> anyhow::Result<()> {
+        // Adding a version?
+        //
+        // 1. Update the LAST_SUPPORTED_UPGRADE_FROM_VERSION and LAST_SUPPORTED_UPGRADE_TO_VERSION.
+        // 2. Add new version to the upgrade list if necessary
+        // 3. Use `no_upgrade` as index for versions that are compatible.
+
+        if self.current_version == self.target_version {
+            println!("Database is already at the target version. Exiting.");
+            return Ok(());
+        }
+
+        if self.current_version > self.target_version {
+            bail!(
+                "Cannot downgrade from {}.{}.{} to {}.{}.{}. Downgrade not supported",
+                self.current_version.0,
+                self.current_version.1,
+                self.current_version.2,
+                self.target_version.0,
+                self.target_version.1,
+                self.target_version.2
+            );
+        }
+
+        const FIRST_SUPPORTED_UPGRADE_FROM_VERSION: &str = "1.9.0";
+        const LAST_SUPPORTED_UPGRADE_FROM_VERSION: &str = "1.12.7";
+        const FIRST_SUPPORTED_UPGRADE_TO_VERSION: &str = "1.10.0";
+        const LAST_SUPPORTED_UPGRADE_TO_VERSION: &str = "1.12.7";
+
         let upgrade_list = [
-            (v1_9_to_v1_10 as fn(&Path) -> Result<(), anyhow::Error>, "1", "10", "0"),
+            (
+                v1_9_to_v1_10 as fn(&Path, &str, &str, &str) -> Result<(), anyhow::Error>,
+                "1",
+                "10",
+                "0",
+            ),
             (v1_10_to_v1_11, "1", "11", "0"),
+            (v1_11_to_v1_12, "1", "12", "0"),
+            (v1_12_to_v1_12_3, "1", "12", "3"),
         ];
 
+        let no_upgrade: usize = upgrade_list.len();
+
         let (current_major, current_minor, current_patch) = &self.current_version;
 
         let start_at = match (
@@ -33,8 +71,13 @@ impl OfflineUpgrade {
         ) {
             ("1", "9", _) => 0,
             ("1", "10", _) => 1,
+            ("1", "11", _) => 2,
+            ("1", "12", "0" | "1" | "2") => 3,
+            ("1", "12", "3" | "4" | "5" | "6" | "7") => no_upgrade,
             _ => {
-                bail!("Unsupported current version {current_major}.{current_minor}.{current_patch}. Can only upgrade from v1.9 and v1.10")
+                bail!("Unsupported current version {current_major}.{current_minor}.{current_patch}. Can only upgrade from versions in range [{}-{}]",
+                      FIRST_SUPPORTED_UPGRADE_FROM_VERSION,
+                      LAST_SUPPORTED_UPGRADE_FROM_VERSION);
             }
         };
 
@@ -43,20 +86,32 @@ impl OfflineUpgrade {
         let ends_at = match (target_major.as_str(), target_minor.as_str(), target_patch.as_str()) {
             ("1", "10", _) => 0,
             ("1", "11", _) => 1,
+            ("1", "12", "0" | "1" | "2") => 2,
+            ("1", "12", "3" | "4" | "5" | "6" | "7") => 3,
             (major, _, _) if major.starts_with('v') => {
                 bail!("Target version must not starts with a `v`. Instead of writing `v1.9.0` write `1.9.0` for example.")
             }
             _ => {
-                bail!("Unsupported target version {target_major}.{target_minor}.{target_patch}. Can only upgrade to v1.10 and v1.11")
+                bail!("Unsupported target version {target_major}.{target_minor}.{target_patch}. Can only upgrade to versions in range [{}-{}]",
+                      FIRST_SUPPORTED_UPGRADE_TO_VERSION,
+                      LAST_SUPPORTED_UPGRADE_TO_VERSION);
             }
         };
 
         println!("Starting the upgrade from {current_major}.{current_minor}.{current_patch} to {target_major}.{target_minor}.{target_patch}");
 
+        if start_at == no_upgrade {
+            println!("No upgrade operation to perform, writing VERSION file");
+            create_version_file(&self.db_path, target_major, target_minor, target_patch)
+                .context("while writing VERSION file after the upgrade")?;
+            println!("Success");
+            return Ok(());
+        }
+
         #[allow(clippy::needless_range_loop)]
         for index in start_at..=ends_at {
             let (func, major, minor, patch) = upgrade_list[index];
-            (func)(&self.db_path)?;
+            (func)(&self.db_path, current_major, current_minor, current_patch)?;
             println!("Done");
             // We're writing the version file just in case an issue arise _while_ upgrading.
             // We don't want the DB to fail in an unknown state.

crates/meilitool/src/upgrade/v1_10.rs

@@ -1,18 +1,13 @@
-use anyhow::bail;
 use std::path::Path;
 
-use anyhow::Context;
-use meilisearch_types::{
-    heed::{
-        types::{SerdeJson, Str},
-        Database, Env, EnvOpenOptions, RoTxn, RwTxn, Unspecified,
-    },
-    milli::index::{db_name, main_key},
-};
-
-use crate::{try_opening_database, try_opening_poly_database, uuid_codec::UuidCodec};
+use anyhow::{bail, Context};
+use meilisearch_types::heed::types::{SerdeJson, Str};
+use meilisearch_types::heed::{Database, Env, EnvOpenOptions, RoTxn, RwTxn, Unspecified};
+use meilisearch_types::milli::index::{db_name, main_key};
 
 use super::v1_9;
+use crate::uuid_codec::UuidCodec;
+use crate::{try_opening_database, try_opening_poly_database};
 
 pub type FieldDistribution = std::collections::BTreeMap<String, u64>;
 
@@ -156,7 +151,12 @@ fn date_round_trip(
     Ok(())
 }
 
-pub fn v1_9_to_v1_10(db_path: &Path) -> anyhow::Result<()> {
+pub fn v1_9_to_v1_10(
+    db_path: &Path,
+    _origin_major: &str,
+    _origin_minor: &str,
+    _origin_patch: &str,
+) -> anyhow::Result<()> {
     println!("Upgrading from v1.9.0 to v1.10.0");
     // 2 changes here
 

crates/meilitool/src/upgrade/v1_11.rs

@@ -7,14 +7,19 @@
 use std::path::Path;
 
 use anyhow::Context;
-use meilisearch_types::{
-    heed::{types::Str, Database, EnvOpenOptions},
-    milli::index::db_name,
-};
+use meilisearch_types::heed::types::Str;
+use meilisearch_types::heed::{Database, EnvOpenOptions};
+use meilisearch_types::milli::index::db_name;
 
-use crate::{try_opening_database, try_opening_poly_database, uuid_codec::UuidCodec};
+use crate::uuid_codec::UuidCodec;
+use crate::{try_opening_database, try_opening_poly_database};
 
-pub fn v1_10_to_v1_11(db_path: &Path) -> anyhow::Result<()> {
+pub fn v1_10_to_v1_11(
+    db_path: &Path,
+    _origin_major: &str,
+    _origin_minor: &str,
+    _origin_patch: &str,
+) -> anyhow::Result<()> {
     println!("Upgrading from v1.10.0 to v1.11.0");
 
     let index_scheduler_path = db_path.join("tasks");

crates/meilitool/src/upgrade/v1_12.rs

@@ -0,0 +1,298 @@
+//! The breaking changes that happened between the v1.11 and the v1.12 are:
+//! - The new indexer changed the update files format from OBKV to ndjson. https://github.com/meilisearch/meilisearch/pull/4900
+
+use std::borrow::Cow;
+use std::io::BufWriter;
+use std::path::Path;
+use std::sync::atomic::AtomicBool;
+
+use anyhow::Context;
+use file_store::FileStore;
+use indexmap::IndexMap;
+use meilisearch_types::milli::documents::DocumentsBatchReader;
+use meilisearch_types::milli::heed::types::{SerdeJson, Str};
+use meilisearch_types::milli::heed::{Database, EnvOpenOptions, RoTxn, RwTxn};
+use meilisearch_types::milli::progress::Step;
+use meilisearch_types::milli::{FieldDistribution, Index};
+use serde::Serialize;
+use serde_json::value::RawValue;
+use tempfile::NamedTempFile;
+use time::OffsetDateTime;
+use uuid::Uuid;
+
+use crate::try_opening_database;
+use crate::uuid_codec::UuidCodec;
+
+pub fn v1_11_to_v1_12(
+    db_path: &Path,
+    _origin_major: &str,
+    _origin_minor: &str,
+    _origin_patch: &str,
+) -> anyhow::Result<()> {
+    println!("Upgrading from v1.11.0 to v1.12.0");
+
+    convert_update_files(db_path)?;
+
+    Ok(())
+}
+
+pub fn v1_12_to_v1_12_3(
+    db_path: &Path,
+    origin_major: &str,
+    origin_minor: &str,
+    origin_patch: &str,
+) -> anyhow::Result<()> {
+    println!("Upgrading from v1.12.{{0, 1, 2}} to v1.12.3");
+
+    if origin_minor == "12" {
+        rebuild_field_distribution(db_path)?;
+    } else {
+        println!("Not rebuilding field distribution as it wasn't corrupted coming from v{origin_major}.{origin_minor}.{origin_patch}");
+    }
+
+    Ok(())
+}
+
+/// Convert the update files from OBKV to ndjson format.
+///
+/// 1) List all the update files using the file store.
+/// 2) For each update file, read the update file into a DocumentsBatchReader.
+/// 3) For each document in the update file, convert the document to a JSON object.
+/// 4) Write the JSON object to a tmp file in the update files directory.
+/// 5) Persist the tmp file replacing the old update file.
+fn convert_update_files(db_path: &Path) -> anyhow::Result<()> {
+    let update_files_dir_path = db_path.join("update_files");
+    let file_store = FileStore::new(&update_files_dir_path).with_context(|| {
+        format!("while creating file store for update files dir {update_files_dir_path:?}")
+    })?;
+
+    for uuid in file_store.all_uuids().context("while retrieving uuids from file store")? {
+        let uuid = uuid.context("while retrieving uuid from file store")?;
+        let update_file_path = file_store.get_update_path(uuid);
+        let update_file = file_store
+            .get_update(uuid)
+            .with_context(|| format!("while getting update file for uuid {uuid:?}"))?;
+
+        let mut file =
+            NamedTempFile::new_in(&update_files_dir_path).map(BufWriter::new).with_context(
+                || format!("while creating bufwriter for update file {update_file_path:?}"),
+            )?;
+
+        let reader = DocumentsBatchReader::from_reader(update_file).with_context(|| {
+            format!("while creating documents batch reader for update file {update_file_path:?}")
+        })?;
+        let (mut cursor, index) = reader.into_cursor_and_fields_index();
+
+        while let Some(document) = cursor.next_document().with_context(|| {
+            format!(
+                "while reading documents from batch reader for update file {update_file_path:?}"
+            )
+        })? {
+            let mut json_document = IndexMap::new();
+            for (fid, value) in document {
+                let field_name = index
+                    .name(fid)
+                    .with_context(|| format!("while getting field name for fid {fid} for update file {update_file_path:?}"))?;
+                let value: &RawValue = serde_json::from_slice(value)?;
+                json_document.insert(field_name, value);
+            }
+
+            serde_json::to_writer(&mut file, &json_document)?;
+        }
+
+        let file = file.into_inner().map_err(|e| e.into_error()).context(format!(
+            "while flushing update file bufwriter for update file {update_file_path:?}"
+        ))?;
+        let _ = file
+            // atomically replace the obkv file with the rewritten NDJSON file
+            .persist(&update_file_path)
+            .with_context(|| format!("while persisting update file {update_file_path:?}"))?;
+    }
+
+    Ok(())
+}
+
+/// Rebuild field distribution as it was wrongly computed in v1.12.x if x < 3
+fn rebuild_field_distribution(db_path: &Path) -> anyhow::Result<()> {
+    let index_scheduler_path = db_path.join("tasks");
+    let env = unsafe { EnvOpenOptions::new().max_dbs(100).open(&index_scheduler_path) }
+        .with_context(|| format!("While trying to open {:?}", index_scheduler_path.display()))?;
+
+    let mut sched_wtxn = env.write_txn()?;
+
+    let index_mapping: Database<Str, UuidCodec> =
+        try_opening_database(&env, &sched_wtxn, "index-mapping")?;
+    let stats_db: Database<UuidCodec, SerdeJson<IndexStats>> =
+        try_opening_database(&env, &sched_wtxn, "index-stats").with_context(|| {
+            format!("While trying to open {:?}", index_scheduler_path.display())
+        })?;
+
+    let index_count =
+        index_mapping.len(&sched_wtxn).context("while reading the number of indexes")?;
+
+    // FIXME: not ideal, we have to pre-populate all indexes to prevent double borrow of sched_wtxn
+    // 1. immutably for the iteration
+    // 2. mutably for updating index stats
+    let indexes: Vec<_> = index_mapping
+        .iter(&sched_wtxn)?
+        .map(|res| res.map(|(uid, uuid)| (uid.to_owned(), uuid)))
+        .collect();
+
+    let progress = meilisearch_types::milli::progress::Progress::default();
+    let finished = AtomicBool::new(false);
+
+    std::thread::scope(|scope| {
+        let display_progress = std::thread::Builder::new()
+            .name("display_progress".into())
+            .spawn_scoped(scope, || {
+                while !finished.load(std::sync::atomic::Ordering::Relaxed) {
+                    std::thread::sleep(std::time::Duration::from_secs(5));
+                    let view = progress.as_progress_view();
+                    let Ok(view) = serde_json::to_string(&view) else {
+                        continue;
+                    };
+                    println!("{view}");
+                }
+            })
+            .unwrap();
+
+        for (index_index, result) in indexes.into_iter().enumerate() {
+            let (uid, uuid) = result?;
+            progress.update_progress(VariableNameStep::new(
+                &uid,
+                index_index as u32,
+                index_count as u32,
+            ));
+            let index_path = db_path.join("indexes").join(uuid.to_string());
+
+            println!(
+                "[{}/{index_count}]Updating index `{uid}` at `{}`",
+                index_index + 1,
+                index_path.display()
+            );
+
+            println!("\t- Rebuilding field distribution");
+
+            let index = meilisearch_types::milli::Index::new(EnvOpenOptions::new(), &index_path)
+                .with_context(|| {
+                    format!("while opening index {uid} at '{}'", index_path.display())
+                })?;
+
+            let mut index_txn = index.write_txn()?;
+
+            meilisearch_types::milli::update::new::reindex::field_distribution(
+                &index,
+                &mut index_txn,
+                &progress,
+            )
+            .context("while rebuilding field distribution")?;
+
+            let stats = IndexStats::new(&index, &index_txn)
+                .with_context(|| format!("computing stats for index `{uid}`"))?;
+            store_stats_of(stats_db, uuid, &mut sched_wtxn, &uid, &stats)?;
+
+            index_txn.commit().context("while committing the write txn for the updated index")?;
+        }
+
+        sched_wtxn.commit().context("while committing the write txn for the index-scheduler")?;
+
+        finished.store(true, std::sync::atomic::Ordering::Relaxed);
+
+        if let Err(panic) = display_progress.join() {
+            let msg = match panic.downcast_ref::<&'static str>() {
+                Some(s) => *s,
+                None => match panic.downcast_ref::<String>() {
+                    Some(s) => &s[..],
+                    None => "Box<dyn Any>",
+                },
+            };
+            eprintln!("WARN: the display thread panicked with {msg}");
+        }
+
+        println!("Upgrading database succeeded");
+        Ok(())
+    })
+}
+
+pub struct VariableNameStep {
+    name: String,
+    current: u32,
+    total: u32,
+}
+
+impl VariableNameStep {
+    pub fn new(name: impl Into<String>, current: u32, total: u32) -> Self {
+        Self { name: name.into(), current, total }
+    }
+}
+
+impl Step for VariableNameStep {
+    fn name(&self) -> Cow<'static, str> {
+        self.name.clone().into()
+    }
+
+    fn current(&self) -> u32 {
+        self.current
+    }
+
+    fn total(&self) -> u32 {
+        self.total
+    }
+}
+
+pub fn store_stats_of(
+    stats_db: Database<UuidCodec, SerdeJson<IndexStats>>,
+    index_uuid: Uuid,
+    sched_wtxn: &mut RwTxn,
+    index_uid: &str,
+    stats: &IndexStats,
+) -> anyhow::Result<()> {
+    stats_db
+        .put(sched_wtxn, &index_uuid, stats)
+        .with_context(|| format!("storing stats for index `{index_uid}`"))?;
+    Ok(())
+}
+
+/// The statistics that can be computed from an `Index` object.
+#[derive(Serialize, Debug)]
+pub struct IndexStats {
+    /// Number of documents in the index.
+    pub number_of_documents: u64,
+    /// Size taken up by the index' DB, in bytes.
+    ///
+    /// This includes the size taken by both the used and free pages of the DB, and as the free pages
+    /// are not returned to the disk after a deletion, this number is typically larger than
+    /// `used_database_size` that only includes the size of the used pages.
+    pub database_size: u64,
+    /// Size taken by the used pages of the index' DB, in bytes.
+    ///
+    /// As the DB backend does not return to the disk the pages that are not currently used by the DB,
+    /// this value is typically smaller than `database_size`.
+    pub used_database_size: u64,
+    /// Association of every field name with the number of times it occurs in the documents.
+    pub field_distribution: FieldDistribution,
+    /// Creation date of the index.
+    #[serde(with = "time::serde::rfc3339")]
+    pub created_at: OffsetDateTime,
+    /// Date of the last update of the index.
+    #[serde(with = "time::serde::rfc3339")]
+    pub updated_at: OffsetDateTime,
+}
+
+impl IndexStats {
+    /// Compute the stats of an index
+    ///
+    /// # Parameters
+    ///
+    /// - rtxn: a RO transaction for the index, obtained from `Index::read_txn()`.
+    pub fn new(index: &Index, rtxn: &RoTxn) -> meilisearch_types::milli::Result<Self> {
+        Ok(IndexStats {
+            number_of_documents: index.number_of_documents(rtxn)?,
+            database_size: index.on_disk_size()?,
+            used_database_size: index.used_size()?,
+            field_distribution: index.field_distribution(rtxn)?,
+            created_at: index.created_at(rtxn)?,
+            updated_at: index.updated_at(rtxn)?,
+        })
+    }
+}

crates/milli/Cargo.toml

@@ -42,7 +42,7 @@ obkv = "0.3.0"
 once_cell = "1.19.0"
 ordered-float = "4.2.1"
 rayon = "1.10.0"
-roaring = { version = "0.10.6", features = ["serde"] }
+roaring = { version = "0.10.7", features = ["serde"] }
 rstar = { version = "0.12.0", features = ["serde"] }
 serde = { version = "1.0.204", features = ["derive"] }
 serde_json = { version = "1.0.120", features = ["preserve_order", "raw_value"] }
@@ -91,13 +91,15 @@ ureq = { version = "2.10.0", features = ["json"] }
 url = "2.5.2"
 rayon-par-bridge = "0.1.0"
 hashbrown = "0.15.0"
-raw-collections = { git = "https://github.com/meilisearch/raw-collections.git", version = "0.1.0" }
 bumpalo = "3.16.0"
+bumparaw-collections = "0.1.2"
 thread_local = "1.1.8"
 allocator-api2 = "0.2.18"
 rustc-hash = "2.0.0"
 uell = "0.1.0"
 enum-iterator = "2.1.0"
+bbqueue = { git = "https://github.com/meilisearch/bbqueue" }
+flume = { version = "0.11.1", default-features = false }
 
 [dev-dependencies]
 mimalloc = { version = "0.1.43", default-features = false }

crates/milli/src/documents/mod.rs

@@ -33,7 +33,7 @@ pub fn obkv_to_object(obkv: &KvReader<FieldId>, index: &DocumentsBatchIndex) ->
             let field_name = index
                 .name(field_id)
                 .ok_or(FieldIdMapMissingEntry::FieldId { field_id, process: "obkv_to_object" })?;
-            let value = serde_json::from_slice(value).map_err(InternalError::SerdeJson)?;
+            let value = serde_json::from_slice(value).map_err(InternalError::SerdeJson).unwrap();
             Ok((field_name.to_string(), value))
         })
         .collect()
@@ -84,7 +84,8 @@ impl DocumentsBatchIndex {
             let key =
                 self.0.get_by_left(&k).ok_or(crate::error::InternalError::DatabaseClosing)?.clone();
             let value = serde_json::from_slice::<serde_json::Value>(v)
-                .map_err(crate::error::InternalError::SerdeJson)?;
+                .map_err(crate::error::InternalError::SerdeJson)
+                .unwrap();
             map.insert(key, value);
         }
 

crates/milli/src/documents/primary_key.rs

@@ -92,7 +92,8 @@ impl<'a> PrimaryKey<'a> {
             PrimaryKey::Flat { name: _, field_id } => match document.get(*field_id) {
                 Some(document_id_bytes) => {
                     let document_id = serde_json::from_slice(document_id_bytes)
-                        .map_err(InternalError::SerdeJson)?;
+                        .map_err(InternalError::SerdeJson)
+                        .unwrap();
                     match validate_document_id_value(document_id) {
                         Ok(document_id) => Ok(Ok(document_id)),
                         Err(user_error) => {
@@ -108,7 +109,8 @@ impl<'a> PrimaryKey<'a> {
                     if let Some(field_id) = fields.id(first_level_name) {
                         if let Some(value_bytes) = document.get(field_id) {
                             let object = serde_json::from_slice(value_bytes)
-                                .map_err(InternalError::SerdeJson)?;
+                                .map_err(InternalError::SerdeJson)
+                                .unwrap();
                             fetch_matching_values(object, right, &mut matching_documents_ids);
 
                             if matching_documents_ids.len() >= 2 {
@@ -151,11 +153,12 @@ impl<'a> PrimaryKey<'a> {
                 };
 
                 let document_id: &RawValue =
-                    serde_json::from_slice(document_id).map_err(InternalError::SerdeJson)?;
+                    serde_json::from_slice(document_id).map_err(InternalError::SerdeJson).unwrap();
 
                 let document_id = document_id
                     .deserialize_any(crate::update::new::indexer::de::DocumentIdVisitor(indexer))
-                    .map_err(InternalError::SerdeJson)?;
+                    .map_err(InternalError::SerdeJson)
+                    .unwrap();
 
                 let external_document_id = match document_id {
                     Ok(document_id) => Ok(document_id),
@@ -173,7 +176,7 @@ impl<'a> PrimaryKey<'a> {
 
                     let Some(value) = document.get(fid) else { continue };
                     let value: &RawValue =
-                        serde_json::from_slice(value).map_err(InternalError::SerdeJson)?;
+                        serde_json::from_slice(value).map_err(InternalError::SerdeJson).unwrap();
                     match match_component(first_level, right, value, indexer, &mut docid) {
                         ControlFlow::Continue(()) => continue,
                         ControlFlow::Break(Ok(_)) => {
@@ -183,7 +186,7 @@ impl<'a> PrimaryKey<'a> {
                             .into())
                         }
                         ControlFlow::Break(Err(err)) => {
-                            return Err(InternalError::SerdeJson(err).into())
+                            panic!("{err}");
                         }
                     }
                 }
@@ -280,7 +283,7 @@ fn starts_with(selector: &str, key: &str) -> bool {
 
 pub fn validate_document_id_str(document_id: &str) -> Option<&str> {
     if document_id.is_empty()
-        || document_id.len() > 512
+        || document_id.len() >= 512
         || !document_id.chars().all(|c| c.is_ascii_alphanumeric() || c == '-' || c == '_')
     {
         None

crates/milli/src/error.rs

@@ -3,6 +3,7 @@ use std::convert::Infallible;
 use std::fmt::Write;
 use std::{io, str};
 
+use bstr::BString;
 use heed::{Error as HeedError, MdbError};
 use rayon::ThreadPoolBuildError;
 use rhai::EvalAltResult;
@@ -62,9 +63,9 @@ pub enum InternalError {
     #[error(transparent)]
     Store(#[from] MdbError),
     #[error("Cannot delete {key:?} from database {database_name}: {error}")]
-    StoreDeletion { database_name: &'static str, key: Vec<u8>, error: heed::Error },
+    StoreDeletion { database_name: &'static str, key: BString, error: heed::Error },
     #[error("Cannot insert {key:?} and value with length {value_length} into database {database_name}: {error}")]
-    StorePut { database_name: &'static str, key: Vec<u8>, value_length: usize, error: heed::Error },
+    StorePut { database_name: &'static str, key: BString, value_length: usize, error: heed::Error },
     #[error(transparent)]
     Utf8(#[from] str::Utf8Error),
     #[error("An indexation process was explicitly aborted")]
@@ -113,7 +114,7 @@ pub enum UserError {
         "Document identifier `{}` is invalid. \
 A document identifier can be of type integer or string, \
 only composed of alphanumeric characters (a-z A-Z 0-9), hyphens (-) and underscores (_), \
-and can not be more than 512 bytes.", .document_id.to_string()
+and can not be more than 511 bytes.", .document_id.to_string()
     )]
     InvalidDocumentId { document_id: Value },
     #[error("Invalid facet distribution, {}", format_invalid_filter_distribution(.invalid_facets_name, .valid_facets_name))]

crates/milli/src/heed_codec/facet/mod.rs

@@ -97,7 +97,7 @@ impl<'a> heed::BytesEncode<'a> for FacetGroupValueCodec {
 
     fn bytes_encode(value: &'a Self::EItem) -> Result<Cow<'a, [u8]>, BoxedError> {
         let mut v = vec![value.size];
-        CboRoaringBitmapCodec::serialize_into(&value.bitmap, &mut v);
+        CboRoaringBitmapCodec::serialize_into_vec(&value.bitmap, &mut v);
         Ok(Cow::Owned(v))
     }
 }

crates/milli/src/heed_codec/roaring_bitmap/cbo_roaring_bitmap_codec.rs

@@ -27,18 +27,27 @@ impl CboRoaringBitmapCodec {
         }
     }
 
-    pub fn serialize_into(roaring: &RoaringBitmap, vec: &mut Vec<u8>) {
+    pub fn serialize_into_vec(roaring: &RoaringBitmap, vec: &mut Vec<u8>) {
+        Self::serialize_into_writer(roaring, vec).unwrap()
+    }
+
+    pub fn serialize_into_writer<W: io::Write>(
+        roaring: &RoaringBitmap,
+        mut writer: W,
+    ) -> io::Result<()> {
         if roaring.len() <= THRESHOLD as u64 {
             // If the number of items (u32s) to encode is less than or equal to the threshold
             // it means that it would weigh the same or less than the RoaringBitmap
             // header, so we directly encode them using ByteOrder instead.
             for integer in roaring {
-                vec.write_u32::<NativeEndian>(integer).unwrap();
+                writer.write_u32::<NativeEndian>(integer)?;
             }
         } else {
             // Otherwise, we use the classic RoaringBitmapCodec that writes a header.
-            roaring.serialize_into(vec).unwrap();
+            roaring.serialize_into(writer)?;
         }
+
+        Ok(())
     }
 
     pub fn deserialize_from(mut bytes: &[u8]) -> io::Result<RoaringBitmap> {
@@ -143,7 +152,7 @@ impl CboRoaringBitmapCodec {
             return Ok(None);
         }
 
-        Self::serialize_into(&previous, buffer);
+        Self::serialize_into_vec(&previous, buffer);
         Ok(Some(&buffer[..]))
     }
 }
@@ -169,7 +178,7 @@ impl heed::BytesEncode<'_> for CboRoaringBitmapCodec {
 
     fn bytes_encode(item: &Self::EItem) -> Result<Cow<'_, [u8]>, BoxedError> {
         let mut vec = Vec::with_capacity(Self::serialized_size(item));
-        Self::serialize_into(item, &mut vec);
+        Self::serialize_into_vec(item, &mut vec);
         Ok(Cow::Owned(vec))
     }
 }

crates/milli/src/index.rs

@@ -1734,6 +1734,7 @@ pub(crate) mod tests {
 
     use crate::error::{Error, InternalError};
     use crate::index::{DEFAULT_MIN_WORD_LEN_ONE_TYPO, DEFAULT_MIN_WORD_LEN_TWO_TYPOS};
+    use crate::progress::Progress;
     use crate::update::new::indexer;
     use crate::update::settings::InnerIndexSettings;
     use crate::update::{
@@ -1810,7 +1811,7 @@ pub(crate) mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                &|_progress| (),
+                Progress::default(),
             )?;
 
             if let Some(error) = operation_stats.into_iter().find_map(|stat| stat.error) {
@@ -1821,6 +1822,7 @@ pub(crate) mod tests {
                 indexer::index(
                     wtxn,
                     &self.inner,
+                    &crate::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     indexer_config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1828,7 +1830,7 @@ pub(crate) mod tests {
                     &document_changes,
                     embedders,
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
             })
             .unwrap()?;
@@ -1900,7 +1902,7 @@ pub(crate) mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                &|_progress| (),
+                Progress::default(),
             )?;
 
             if let Some(error) = operation_stats.into_iter().find_map(|stat| stat.error) {
@@ -1911,6 +1913,7 @@ pub(crate) mod tests {
                 indexer::index(
                     wtxn,
                     &self.inner,
+                    &crate::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     indexer_config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1918,7 +1921,7 @@ pub(crate) mod tests {
                     &document_changes,
                     embedders,
                     &|| false,
-                    &|_| (),
+                    &Progress::default(),
                 )
             })
             .unwrap()?;
@@ -1980,7 +1983,7 @@ pub(crate) mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                &|_progress| (),
+                Progress::default(),
             )
             .unwrap();
 
@@ -1991,6 +1994,7 @@ pub(crate) mod tests {
                 indexer::index(
                     &mut wtxn,
                     &index.inner,
+                    &crate::ThreadPoolNoAbortBuilder::new().build().unwrap(),
                     indexer_config.grenad_parameters(),
                     &db_fields_ids_map,
                     new_fields_ids_map,
@@ -1998,7 +2002,7 @@ pub(crate) mod tests {
                     &document_changes,
                     embedders,
                     &|| should_abort.load(Relaxed),
-                    &|_| (),
+                    &Progress::default(),
                 )
             })
             .unwrap()

crates/milli/src/lib.rs

@@ -1,6 +1,7 @@
 #![cfg_attr(all(test, fuzzing), feature(no_coverage))]
 #![allow(clippy::type_complexity)]
 
+#[cfg(not(windows))]
 #[cfg(test)]
 #[global_allocator]
 pub static ALLOC: mimalloc::MiMalloc = mimalloc::MiMalloc;
@@ -30,6 +31,7 @@ pub mod vector;
 #[macro_use]
 pub mod snapshot_tests;
 mod fieldids_weights_map;
+pub mod progress;
 
 use std::collections::{BTreeMap, HashMap};
 use std::convert::{TryFrom, TryInto};
@@ -226,7 +228,8 @@ pub fn obkv_to_json(
                 field_id: id,
                 process: "obkv_to_json",
             })?;
-            let value = serde_json::from_slice(value).map_err(error::InternalError::SerdeJson)?;
+            let value =
+                serde_json::from_slice(value).map_err(error::InternalError::SerdeJson).unwrap();
             Ok((name.to_owned(), value))
         })
         .collect()

crates/milli/src/progress.rs

@@ -0,0 +1,152 @@
+use std::any::TypeId;
+use std::borrow::Cow;
+use std::sync::atomic::{AtomicU32, Ordering};
+use std::sync::{Arc, RwLock};
+
+use serde::Serialize;
+
+pub trait Step: 'static + Send + Sync {
+    fn name(&self) -> Cow<'static, str>;
+    fn current(&self) -> u32;
+    fn total(&self) -> u32;
+}
+
+#[derive(Clone, Default)]
+pub struct Progress {
+    steps: Arc<RwLock<Vec<(TypeId, Box<dyn Step>)>>>,
+}
+
+impl Progress {
+    pub fn update_progress<P: Step>(&self, sub_progress: P) {
+        let mut steps = self.steps.write().unwrap();
+        let step_type = TypeId::of::<P>();
+        if let Some(idx) = steps.iter().position(|(id, _)| *id == step_type) {
+            steps.truncate(idx);
+        }
+        steps.push((step_type, Box::new(sub_progress)));
+    }
+
+    // TODO: This code should be in meilisearch_types but cannot because milli can't depend on meilisearch_types
+    pub fn as_progress_view(&self) -> ProgressView {
+        let steps = self.steps.read().unwrap();
+
+        let mut percentage = 0.0;
+        let mut prev_factors = 1.0;
+
+        let mut step_view = Vec::with_capacity(steps.len());
+        for (_, step) in steps.iter() {
+            prev_factors *= step.total() as f32;
+            percentage += step.current() as f32 / prev_factors;
+
+            step_view.push(ProgressStepView {
+                current_step: step.name(),
+                finished: step.current(),
+                total: step.total(),
+            });
+        }
+
+        ProgressView { steps: step_view, percentage: percentage * 100.0 }
+    }
+}
+
+/// This trait lets you use the AtomicSubStep defined right below.
+/// The name must be a const that never changed but that can't be enforced by the type system because it make the trait non object-safe.
+/// By forcing the Default trait + the &'static str we make it harder to miss-use the trait.
+pub trait NamedStep: 'static + Send + Sync + Default {
+    fn name(&self) -> &'static str;
+}
+
+/// Structure to quickly define steps that need very quick, lockless updating of their current step.
+/// You can use this struct if:
+/// - The name of the step doesn't change
+/// - The total number of steps doesn't change
+pub struct AtomicSubStep<Name: NamedStep> {
+    unit_name: Name,
+    current: Arc<AtomicU32>,
+    total: u32,
+}
+
+impl<Name: NamedStep> AtomicSubStep<Name> {
+    pub fn new(total: u32) -> (Arc<AtomicU32>, Self) {
+        let current = Arc::new(AtomicU32::new(0));
+        (current.clone(), Self { current, total, unit_name: Name::default() })
+    }
+}
+
+impl<Name: NamedStep> Step for AtomicSubStep<Name> {
+    fn name(&self) -> Cow<'static, str> {
+        self.unit_name.name().into()
+    }
+
+    fn current(&self) -> u32 {
+        self.current.load(Ordering::Relaxed)
+    }
+
+    fn total(&self) -> u32 {
+        self.total
+    }
+}
+
+#[macro_export]
+macro_rules! make_enum_progress {
+    ($visibility:vis enum $name:ident { $($variant:ident,)+ }) => {
+        #[repr(u8)]
+        #[derive(Debug, Clone, Copy, PartialEq, Eq, Sequence)]
+        #[allow(clippy::enum_variant_names)]
+        $visibility enum $name {
+            $($variant),+
+        }
+
+        impl Step for $name {
+            fn name(&self) -> Cow<'static, str> {
+                use convert_case::Casing;
+
+                match self {
+                    $(
+                        $name::$variant => stringify!($variant).from_case(convert_case::Case::Camel).to_case(convert_case::Case::Lower).into()
+                    ),+
+                }
+            }
+
+            fn current(&self) -> u32 {
+                *self as u32
+            }
+
+            fn total(&self) -> u32 {
+                Self::CARDINALITY as u32
+            }
+        }
+    };
+}
+
+#[macro_export]
+macro_rules! make_atomic_progress {
+    ($struct_name:ident alias $atomic_struct_name:ident => $step_name:literal) => {
+        #[derive(Default, Debug, Clone, Copy)]
+        pub struct $struct_name {}
+        impl NamedStep for $struct_name {
+            fn name(&self) -> &'static str {
+                $step_name
+            }
+        }
+        pub type $atomic_struct_name = AtomicSubStep<$struct_name>;
+    };
+}
+
+make_atomic_progress!(Document alias AtomicDocumentStep => "document" );
+make_atomic_progress!(Payload alias AtomicPayloadStep => "payload" );
+
+#[derive(Debug, Serialize, Clone)]
+#[serde(rename_all = "camelCase")]
+pub struct ProgressView {
+    pub steps: Vec<ProgressStepView>,
+    pub percentage: f32,
+}
+
+#[derive(Debug, Serialize, Clone)]
+#[serde(rename_all = "camelCase")]
+pub struct ProgressStepView {
+    pub current_step: Cow<'static, str>,
+    pub finished: u32,
+    pub total: u32,
+}

crates/milli/src/prompt/document.rs

@@ -3,12 +3,13 @@ use std::collections::BTreeMap;
 use std::fmt::{self, Debug};
 
 use bumpalo::Bump;
+use bumparaw_collections::{RawMap, RawVec, Value};
 use liquid::model::{
     ArrayView, DisplayCow, KString, KStringCow, ObjectRender, ObjectSource, ScalarCow, State,
     Value as LiquidValue,
 };
 use liquid::{ObjectView, ValueView};
-use raw_collections::{RawMap, RawVec};
+use rustc_hash::FxBuildHasher;
 use serde_json::value::RawValue;
 
 use crate::update::del_add::{DelAdd, KvReaderDelAdd};
@@ -195,7 +196,7 @@ impl<'doc, D: DocumentTrait<'doc> + Debug> ObjectView for ParseableDocument<'doc
 }
 
 impl<'doc, D: DocumentTrait<'doc> + Debug> ValueView for ParseableDocument<'doc, D> {
-    fn as_debug(&self) -> &dyn fmt::Debug {
+    fn as_debug(&self) -> &dyn Debug {
         self
     }
     fn render(&self) -> liquid::model::DisplayCow<'_> {
@@ -243,14 +244,13 @@ impl<'doc, D: DocumentTrait<'doc> + Debug> ValueView for ParseableDocument<'doc,
     }
 }
 
-#[derive(Debug)]
 struct ParseableValue<'doc> {
-    value: raw_collections::Value<'doc>,
+    value: Value<'doc, FxBuildHasher>,
 }
 
 impl<'doc> ParseableValue<'doc> {
     pub fn new(value: &'doc RawValue, doc_alloc: &'doc Bump) -> Self {
-        let value = raw_collections::Value::from_raw_value(value, doc_alloc).unwrap();
+        let value = Value::from_raw_value_and_hasher(value, FxBuildHasher, doc_alloc).unwrap();
         Self { value }
     }
 
@@ -260,19 +260,19 @@ impl<'doc> ParseableValue<'doc> {
 }
 
 // transparent newtype for implementing ValueView
-#[repr(transparent)]
 #[derive(Debug)]
-struct ParseableMap<'doc>(RawMap<'doc>);
+#[repr(transparent)]
+struct ParseableMap<'doc>(RawMap<'doc, FxBuildHasher>);
 
 // transparent newtype for implementing ValueView
-#[repr(transparent)]
 #[derive(Debug)]
+#[repr(transparent)]
 struct ParseableArray<'doc>(RawVec<'doc>);
 
 impl<'doc> ParseableMap<'doc> {
-    pub fn as_parseable<'a>(map: &'a RawMap<'doc>) -> &'a ParseableMap<'doc> {
+    pub fn as_parseable<'a>(map: &'a RawMap<'doc, FxBuildHasher>) -> &'a ParseableMap<'doc> {
         // SAFETY: repr(transparent)
-        unsafe { &*(map as *const RawMap as *const Self) }
+        unsafe { &*(map as *const RawMap<FxBuildHasher> as *const Self) }
     }
 }
 
@@ -447,8 +447,9 @@ impl<'doc> ValueView for ParseableValue<'doc> {
     }
 
     fn render(&self) -> DisplayCow<'_> {
-        use raw_collections::value::Number;
-        use raw_collections::Value;
+        use bumparaw_collections::value::Number;
+        use bumparaw_collections::Value;
+
         match &self.value {
             Value::Null => LiquidValue::Nil.render(),
             Value::Bool(v) => v.render(),
@@ -464,8 +465,9 @@ impl<'doc> ValueView for ParseableValue<'doc> {
     }
 
     fn source(&self) -> DisplayCow<'_> {
-        use raw_collections::value::Number;
-        use raw_collections::Value;
+        use bumparaw_collections::value::Number;
+        use bumparaw_collections::Value;
+
         match &self.value {
             Value::Null => LiquidValue::Nil.source(),
             Value::Bool(v) => ValueView::source(v),
@@ -481,8 +483,9 @@ impl<'doc> ValueView for ParseableValue<'doc> {
     }
 
     fn type_name(&self) -> &'static str {
-        use raw_collections::value::Number;
-        use raw_collections::Value;
+        use bumparaw_collections::value::Number;
+        use bumparaw_collections::Value;
+
         match &self.value {
             Value::Null => LiquidValue::Nil.type_name(),
             Value::Bool(v) => v.type_name(),
@@ -498,7 +501,8 @@ impl<'doc> ValueView for ParseableValue<'doc> {
     }
 
     fn query_state(&self, state: State) -> bool {
-        use raw_collections::Value;
+        use bumparaw_collections::Value;
+
         match &self.value {
             Value::Null => ValueView::query_state(&LiquidValue::Nil, state),
             Value::Bool(v) => ValueView::query_state(v, state),
@@ -515,7 +519,8 @@ impl<'doc> ValueView for ParseableValue<'doc> {
     }
 
     fn to_kstr(&self) -> KStringCow<'_> {
-        use raw_collections::Value;
+        use bumparaw_collections::Value;
+
         match &self.value {
             Value::Null => ValueView::to_kstr(&LiquidValue::Nil),
             Value::Bool(v) => ValueView::to_kstr(v),
@@ -527,12 +532,14 @@ impl<'doc> ValueView for ParseableValue<'doc> {
     }
 
     fn to_value(&self) -> LiquidValue {
-        use raw_collections::Value;
+        use bumparaw_collections::value::Number;
+        use bumparaw_collections::Value;
+
         match &self.value {
             Value::Null => LiquidValue::Nil,
             Value::Bool(v) => LiquidValue::Scalar(liquid::model::ScalarCow::new(*v)),
             Value::Number(number) => match number {
-                raw_collections::value::Number::PosInt(number) => {
+                Number::PosInt(number) => {
                     let number: i64 = match (*number).try_into() {
                         Ok(number) => number,
                         Err(_) => {
@@ -541,12 +548,8 @@ impl<'doc> ValueView for ParseableValue<'doc> {
                     };
                     LiquidValue::Scalar(ScalarCow::new(number))
                 }
-                raw_collections::value::Number::NegInt(number) => {
-                    LiquidValue::Scalar(ScalarCow::new(*number))
-                }
-                raw_collections::value::Number::Finite(number) => {
-                    LiquidValue::Scalar(ScalarCow::new(*number))
-                }
+                Number::NegInt(number) => LiquidValue::Scalar(ScalarCow::new(*number)),
+                Number::Finite(number) => LiquidValue::Scalar(ScalarCow::new(*number)),
             },
             Value::String(s) => LiquidValue::Scalar(liquid::model::ScalarCow::new(s.to_string())),
             Value::Array(raw_vec) => ParseableArray::as_parseable(raw_vec).to_value(),
@@ -555,8 +558,9 @@ impl<'doc> ValueView for ParseableValue<'doc> {
     }
 
     fn as_scalar(&self) -> Option<liquid::model::ScalarCow<'_>> {
-        use raw_collections::value::Number;
-        use raw_collections::Value;
+        use bumparaw_collections::value::Number;
+        use bumparaw_collections::Value;
+
         match &self.value {
             Value::Bool(v) => Some(liquid::model::ScalarCow::new(*v)),
             Value::Number(number) => match number {
@@ -576,34 +580,41 @@ impl<'doc> ValueView for ParseableValue<'doc> {
     }
 
     fn is_scalar(&self) -> bool {
-        use raw_collections::Value;
+        use bumparaw_collections::Value;
+
         matches!(&self.value, Value::Bool(_) | Value::Number(_) | Value::String(_))
     }
 
     fn as_array(&self) -> Option<&dyn liquid::model::ArrayView> {
-        if let raw_collections::Value::Array(array) = &self.value {
+        if let Value::Array(array) = &self.value {
             return Some(ParseableArray::as_parseable(array) as _);
         }
         None
     }
 
     fn is_array(&self) -> bool {
-        matches!(&self.value, raw_collections::Value::Array(_))
+        matches!(&self.value, bumparaw_collections::Value::Array(_))
     }
 
     fn as_object(&self) -> Option<&dyn ObjectView> {
-        if let raw_collections::Value::Object(object) = &self.value {
+        if let Value::Object(object) = &self.value {
             return Some(ParseableMap::as_parseable(object) as _);
         }
         None
     }
 
     fn is_object(&self) -> bool {
-        matches!(&self.value, raw_collections::Value::Object(_))
+        matches!(&self.value, bumparaw_collections::Value::Object(_))
     }
 
     fn is_nil(&self) -> bool {
-        matches!(&self.value, raw_collections::Value::Null)
+        matches!(&self.value, bumparaw_collections::Value::Null)
+    }
+}
+
+impl Debug for ParseableValue<'_> {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        f.debug_struct("ParseableValue").field("value", &self.value).finish()
     }
 }
 

crates/milli/src/prompt/error.rs

@@ -38,6 +38,16 @@ pub struct RenderPromptError {
     pub fault: FaultSource,
 }
 impl RenderPromptError {
+    pub(crate) fn missing_context_with_external_docid(
+        external_docid: String,
+        inner: liquid::Error,
+    ) -> RenderPromptError {
+        Self {
+            kind: RenderPromptErrorKind::MissingContextWithExternalDocid(external_docid, inner),
+            fault: FaultSource::User,
+        }
+    }
+
     pub(crate) fn missing_context(inner: liquid::Error) -> RenderPromptError {
         Self { kind: RenderPromptErrorKind::MissingContext(inner), fault: FaultSource::User }
     }
@@ -47,6 +57,8 @@ impl RenderPromptError {
 pub enum RenderPromptErrorKind {
     #[error("missing field in document: {0}")]
     MissingContext(liquid::Error),
+    #[error("missing field in document `{0}`: {1}")]
+    MissingContextWithExternalDocid(String, liquid::Error),
 }
 
 impl From<RenderPromptError> for crate::Error {

crates/milli/src/prompt/mod.rs

@@ -119,6 +119,7 @@ impl Prompt {
         'doc: 'a, // lifetime of the allocator, will live for an entire chunk of documents
     >(
         &self,
+        external_docid: &str,
         document: impl crate::update::new::document::Document<'a> + Debug,
         field_id_map: &RefCell<GlobalFieldsIdsMap>,
         doc_alloc: &'doc Bump,
@@ -130,9 +131,12 @@ impl Prompt {
             self.max_bytes.unwrap_or_else(default_max_bytes).get(),
             doc_alloc,
         );
-        self.template
-            .render_to(&mut rendered, &context)
-            .map_err(RenderPromptError::missing_context)?;
+        self.template.render_to(&mut rendered, &context).map_err(|liquid_error| {
+            RenderPromptError::missing_context_with_external_docid(
+                external_docid.to_owned(),
+                liquid_error,
+            )
+        })?;
         Ok(std::str::from_utf8(rendered.into_bump_slice())
             .expect("render can only write UTF-8 because all inputs and processing preserve utf-8"))
     }

crates/milli/src/search/facet/facet_distribution.rs

@@ -219,12 +219,19 @@ impl<'a> FacetDistribution<'a> {
                 let facet_key = StrRefCodec::bytes_decode(facet_key).unwrap();
 
                 let key: (FieldId, _, &str) = (field_id, any_docid, facet_key);
-                let original_string = self
-                    .index
-                    .field_id_docid_facet_strings
-                    .get(self.rtxn, &key)?
-                    .unwrap()
-                    .to_owned();
+                let optional_original_string =
+                    self.index.field_id_docid_facet_strings.get(self.rtxn, &key)?;
+
+                let original_string = match optional_original_string {
+                    Some(original_string) => original_string.to_owned(),
+                    None => {
+                        tracing::error!(
+                            "Missing original facet string. Using the normalized facet {} instead",
+                            facet_key
+                        );
+                        facet_key.to_string()
+                    }
+                };
 
                 distribution.insert(original_string, nbr_docids);
                 if distribution.len() == self.max_values_per_facet {

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

@@ -5,6 +5,7 @@ use bumpalo::Bump;
 use heed::EnvOpenOptions;
 use maplit::{btreemap, hashset};
 
+use crate::progress::Progress;
 use crate::update::new::indexer;
 use crate::update::{IndexDocumentsMethod, IndexerConfig, Settings};
 use crate::vector::EmbeddingConfigs;
@@ -72,7 +73,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
             None,
             &mut new_fields_ids_map,
             &|| false,
-            &|_progress| (),
+            Progress::default(),
         )
         .unwrap();
 
@@ -83,6 +84,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
     indexer::index(
         &mut wtxn,
         &index,
+        &crate::ThreadPoolNoAbortBuilder::new().build().unwrap(),
         config.grenad_parameters(),
         &db_fields_ids_map,
         new_fields_ids_map,
@@ -90,7 +92,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
         &document_changes,
         embedders,
         &|| false,
-        &|_| (),
+        &Progress::default(),
     )
     .unwrap();
 

crates/milli/src/thread_pool_no_abort.rs

@@ -1,4 +1,4 @@
-use std::sync::atomic::{AtomicBool, Ordering};
+use std::sync::atomic::{AtomicBool, AtomicUsize, Ordering};
 use std::sync::Arc;
 
 use rayon::{ThreadPool, ThreadPoolBuilder};
@@ -9,6 +9,8 @@ use thiserror::Error;
 #[derive(Debug)]
 pub struct ThreadPoolNoAbort {
     thread_pool: ThreadPool,
+    /// The number of active operations.
+    active_operations: AtomicUsize,
     /// Set to true if the thread pool catched a panic.
     pool_catched_panic: Arc<AtomicBool>,
 }
@@ -19,7 +21,9 @@ impl ThreadPoolNoAbort {
         OP: FnOnce() -> R + Send,
         R: Send,
     {
+        self.active_operations.fetch_add(1, Ordering::Relaxed);
         let output = self.thread_pool.install(op);
+        self.active_operations.fetch_sub(1, Ordering::Relaxed);
         // While reseting the pool panic catcher we return an error if we catched one.
         if self.pool_catched_panic.swap(false, Ordering::SeqCst) {
             Err(PanicCatched)
@@ -31,6 +35,11 @@ impl ThreadPoolNoAbort {
     pub fn current_num_threads(&self) -> usize {
         self.thread_pool.current_num_threads()
     }
+
+    /// The number of active operations.
+    pub fn active_operations(&self) -> usize {
+        self.active_operations.load(Ordering::Relaxed)
+    }
 }
 
 #[derive(Error, Debug)]
@@ -64,6 +73,10 @@ impl ThreadPoolNoAbortBuilder {
             let catched_panic = pool_catched_panic.clone();
             move |_result| catched_panic.store(true, Ordering::SeqCst)
         });
-        Ok(ThreadPoolNoAbort { thread_pool: self.0.build()?, pool_catched_panic })
+        Ok(ThreadPoolNoAbort {
+            thread_pool: self.0.build()?,
+            active_operations: AtomicUsize::new(0),
+            pool_catched_panic,
+        })
     }
 }

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

@@ -79,22 +79,29 @@ pub const FACET_MIN_LEVEL_SIZE: u8 = 5;
 use std::collections::BTreeSet;
 use std::fs::File;
 use std::io::BufReader;
+use std::ops::Bound;
 
 use grenad::Merger;
 use heed::types::{Bytes, DecodeIgnore};
+use heed::BytesDecode as _;
+use roaring::RoaringBitmap;
 use time::OffsetDateTime;
 use tracing::debug;
 
 use self::incremental::FacetsUpdateIncremental;
 use super::{FacetsUpdateBulk, MergeDeladdBtreesetString, MergeDeladdCboRoaringBitmaps};
 use crate::facet::FacetType;
-use crate::heed_codec::facet::{FacetGroupKey, FacetGroupKeyCodec, FacetGroupValueCodec};
+use crate::heed_codec::facet::{
+    FacetGroupKey, FacetGroupKeyCodec, FacetGroupValueCodec, OrderedF64Codec,
+};
 use crate::heed_codec::BytesRefCodec;
+use crate::search::facet::get_highest_level;
 use crate::update::del_add::{DelAdd, KvReaderDelAdd};
 use crate::{try_split_array_at, FieldId, Index, Result};
 
 pub mod bulk;
 pub mod incremental;
+pub mod new_incremental;
 
 /// A builder used to add new elements to the `facet_id_string_docids` or `facet_id_f64_docids` databases.
 ///
@@ -646,3 +653,194 @@ mod comparison_bench {
         }
     }
 }
+
+/// Run sanity checks on the specified fid tree
+///
+/// 1. No "orphan" child value, any child value has a parent
+/// 2. Any docid in the child appears in the parent
+/// 3. No docid in the parent is missing from all its children
+/// 4. no group is bigger than max_group_size
+/// 5. Less than 50% of groups are bigger than group_size
+/// 6. group size matches the number of children
+/// 7. max_level is < 255
+pub(crate) fn sanity_checks(
+    index: &Index,
+    rtxn: &heed::RoTxn,
+    field_id: FieldId,
+    facet_type: FacetType,
+    group_size: usize,
+    _min_level_size: usize, // might add a check on level size later
+    max_group_size: usize,
+) -> Result<()> {
+    tracing::info!(%field_id, ?facet_type, "performing sanity checks");
+    let database = match facet_type {
+        FacetType::String => {
+            index.facet_id_string_docids.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>()
+        }
+        FacetType::Number => {
+            index.facet_id_f64_docids.remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>()
+        }
+    };
+
+    let leaf_prefix: FacetGroupKey<&[u8]> = FacetGroupKey { field_id, level: 0, left_bound: &[] };
+
+    let leaf_it = database.prefix_iter(rtxn, &leaf_prefix)?;
+
+    let max_level = get_highest_level(rtxn, database, field_id)?;
+    if max_level == u8::MAX {
+        panic!("max_level == 255");
+    }
+
+    for leaf in leaf_it {
+        let (leaf_facet_value, leaf_docids) = leaf?;
+        let mut current_level = 0;
+
+        let mut current_parent_facet_value: Option<FacetGroupKey<&[u8]>> = None;
+        let mut current_parent_docids: Option<crate::heed_codec::facet::FacetGroupValue> = None;
+        loop {
+            current_level += 1;
+            if current_level >= max_level {
+                break;
+            }
+            let parent_key_right_bound = FacetGroupKey {
+                field_id,
+                level: current_level,
+                left_bound: leaf_facet_value.left_bound,
+            };
+            let (parent_facet_value, parent_docids) = database
+                .get_lower_than_or_equal_to(rtxn, &parent_key_right_bound)?
+                .expect("no parent found");
+            if parent_facet_value.level != current_level {
+                panic!(
+                    "wrong parent level, found_level={}, expected_level={}",
+                    parent_facet_value.level, current_level
+                );
+            }
+            if parent_facet_value.field_id != field_id {
+                panic!("wrong parent fid");
+            }
+            if parent_facet_value.left_bound > leaf_facet_value.left_bound {
+                panic!("wrong parent left bound");
+            }
+
+            if !leaf_docids.bitmap.is_subset(&parent_docids.bitmap) {
+                panic!(
+                    "missing docids from leaf in parent, current_level={}, parent={}, child={}, missing={missing:?}, child_len={}, child={:?}",
+                    current_level,
+                    facet_to_string(parent_facet_value.left_bound, facet_type),
+                    facet_to_string(leaf_facet_value.left_bound, facet_type),
+                    leaf_docids.bitmap.len(),
+                    leaf_docids.bitmap.clone(),
+                    missing=leaf_docids.bitmap - parent_docids.bitmap,
+                )
+            }
+
+            if let Some(current_parent_facet_value) = current_parent_facet_value {
+                if current_parent_facet_value.field_id != parent_facet_value.field_id {
+                    panic!("wrong parent parent fid");
+                }
+                if current_parent_facet_value.level + 1 != parent_facet_value.level {
+                    panic!("wrong parent parent level");
+                }
+                if current_parent_facet_value.left_bound < parent_facet_value.left_bound {
+                    panic!("wrong parent parent left bound");
+                }
+            }
+
+            if let Some(current_parent_docids) = current_parent_docids {
+                if !current_parent_docids.bitmap.is_subset(&parent_docids.bitmap) {
+                    panic!("missing docids from intermediate node in parent, parent_level={}, parent={}, intermediate={}, missing={missing:?}, intermediate={:?}",
+                    parent_facet_value.level,
+                    facet_to_string(parent_facet_value.left_bound, facet_type),
+                    facet_to_string(current_parent_facet_value.unwrap().left_bound, facet_type),
+                    current_parent_docids.bitmap.clone(),
+                    missing=current_parent_docids.bitmap - parent_docids.bitmap,
+                    );
+                }
+            }
+
+            current_parent_facet_value = Some(parent_facet_value);
+            current_parent_docids = Some(parent_docids);
+        }
+    }
+    tracing::info!(%field_id, ?facet_type, "checked all leaves");
+
+    let mut current_level = max_level;
+    let mut greater_than_group = 0usize;
+    let mut total = 0usize;
+    loop {
+        if current_level == 0 {
+            break;
+        }
+        let child_level = current_level - 1;
+        tracing::info!(%field_id, ?facet_type, %current_level, "checked groups for level");
+        let level_groups_prefix: FacetGroupKey<&[u8]> =
+            FacetGroupKey { field_id, level: current_level, left_bound: &[] };
+        let mut level_groups_it = database.prefix_iter(rtxn, &level_groups_prefix)?.peekable();
+
+        'group_it: loop {
+            let Some(group) = level_groups_it.next() else { break 'group_it };
+
+            let (group_facet_value, group_docids) = group?;
+            let child_left_bound = group_facet_value.left_bound.to_owned();
+            let mut expected_docids = RoaringBitmap::new();
+            let mut expected_size = 0usize;
+            let right_bound = level_groups_it
+                .peek()
+                .and_then(|res| res.as_ref().ok())
+                .map(|(key, _)| key.left_bound);
+            let child_left_bound = FacetGroupKey {
+                field_id,
+                level: child_level,
+                left_bound: child_left_bound.as_slice(),
+            };
+            let child_left_bound = Bound::Included(&child_left_bound);
+            let child_right_bound;
+            let child_right_bound = if let Some(right_bound) = right_bound {
+                child_right_bound =
+                    FacetGroupKey { field_id, level: child_level, left_bound: right_bound };
+                Bound::Excluded(&child_right_bound)
+            } else {
+                Bound::Unbounded
+            };
+            let children = database.range(rtxn, &(child_left_bound, child_right_bound))?;
+            for child in children {
+                let (child_facet_value, child_docids) = child?;
+                if child_facet_value.field_id != field_id {
+                    break;
+                }
+                if child_facet_value.level != child_level {
+                    break;
+                }
+                expected_size += 1;
+                expected_docids |= &child_docids.bitmap;
+            }
+            assert_eq!(expected_size, group_docids.size as usize);
+            assert!(expected_size <= max_group_size);
+            assert_eq!(expected_docids, group_docids.bitmap);
+            total += 1;
+            if expected_size > group_size {
+                greater_than_group += 1;
+            }
+        }
+
+        current_level -= 1;
+    }
+    if greater_than_group * 2 > total {
+        panic!("too many groups have a size > group_size");
+    }
+
+    tracing::info!("sanity checks OK");
+
+    Ok(())
+}
+
+fn facet_to_string(facet_value: &[u8], facet_type: FacetType) -> String {
+    match facet_type {
+        FacetType::String => bstr::BStr::new(facet_value).to_string(),
+        FacetType::Number => match OrderedF64Codec::bytes_decode(facet_value) {
+            Ok(value) => value.to_string(),
+            Err(e) => format!("error: {e} (bytes: {facet_value:?}"),
+        },
+    }
+}

crates/milli/src/update/facet/new_incremental.rs

@@ -0,0 +1,498 @@
+use std::ops::Bound;
+
+use heed::types::{Bytes, DecodeIgnore};
+use heed::{BytesDecode as _, Database, RwTxn};
+use roaring::RoaringBitmap;
+
+use crate::facet::FacetType;
+use crate::heed_codec::facet::{
+    FacetGroupKey, FacetGroupKeyCodec, FacetGroupValue, FacetGroupValueCodec,
+};
+use crate::heed_codec::BytesRefCodec;
+use crate::search::facet::get_highest_level;
+use crate::update::valid_facet_value;
+use crate::{FieldId, Index, Result};
+
+pub struct FacetsUpdateIncremental {
+    inner: FacetsUpdateIncrementalInner,
+    delta_data: Vec<FacetFieldIdChange>,
+}
+
+struct FacetsUpdateIncrementalInner {
+    db: Database<FacetGroupKeyCodec<BytesRefCodec>, FacetGroupValueCodec>,
+    field_id: FieldId,
+    group_size: u8,
+    min_level_size: u8,
+    max_group_size: u8,
+}
+
+impl FacetsUpdateIncremental {
+    pub fn new(
+        index: &Index,
+        facet_type: FacetType,
+        field_id: FieldId,
+        delta_data: Vec<FacetFieldIdChange>,
+        group_size: u8,
+        min_level_size: u8,
+        max_group_size: u8,
+    ) -> Self {
+        FacetsUpdateIncremental {
+            inner: FacetsUpdateIncrementalInner {
+                db: match facet_type {
+                    FacetType::String => index
+                        .facet_id_string_docids
+                        .remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>(),
+                    FacetType::Number => index
+                        .facet_id_f64_docids
+                        .remap_key_type::<FacetGroupKeyCodec<BytesRefCodec>>(),
+                },
+                field_id,
+                group_size,
+                min_level_size,
+                max_group_size,
+            },
+
+            delta_data,
+        }
+    }
+
+    #[tracing::instrument(level = "trace", skip_all, target = "indexing::facets::incremental")]
+    pub fn execute(mut self, wtxn: &mut RwTxn) -> Result<()> {
+        if self.delta_data.is_empty() {
+            return Ok(());
+        }
+        self.delta_data.sort_unstable_by(
+            |FacetFieldIdChange { facet_value: left, .. },
+             FacetFieldIdChange { facet_value: right, .. }| {
+                left.cmp(right)
+                    // sort in **reverse** lexicographic order
+                    .reverse()
+            },
+        );
+
+        self.inner.find_changed_parents(wtxn, self.delta_data)?;
+
+        self.inner.add_or_delete_level(wtxn)
+    }
+}
+
+impl FacetsUpdateIncrementalInner {
+    /// WARNING: `changed_children` must be sorted in **reverse** lexicographic order.
+    fn find_changed_parents(
+        &self,
+        wtxn: &mut RwTxn,
+        mut changed_children: Vec<FacetFieldIdChange>,
+    ) -> Result<()> {
+        let mut changed_parents = vec![];
+        for child_level in 0u8..u8::MAX {
+            // child_level < u8::MAX by construction
+            let parent_level = child_level + 1;
+            let parent_level_left_bound: FacetGroupKey<&[u8]> =
+                FacetGroupKey { field_id: self.field_id, level: parent_level, left_bound: &[] };
+
+            let mut last_parent: Option<Box<[u8]>> = None;
+            let mut child_it = changed_children
+                // drain all changed children
+                .drain(..)
+                // keep only children whose value is valid in the LMDB sense
+                .filter(|child| valid_facet_value(&child.facet_value));
+            // `while let` rather than `for` because we advance `child_it` inside of the loop
+            'current_level: while let Some(child) = child_it.next() {
+                if let Some(last_parent) = &last_parent {
+                    if &child.facet_value >= last_parent {
+                        self.compute_parent_group(wtxn, child_level, child.facet_value)?;
+                        continue 'current_level;
+                    }
+                }
+
+                // need to find a new parent
+                let parent_key_prefix = FacetGroupKey {
+                    field_id: self.field_id,
+                    level: parent_level,
+                    left_bound: &*child.facet_value,
+                };
+
+                let parent = self
+                    .db
+                    .remap_data_type::<DecodeIgnore>()
+                    .rev_range(
+                        wtxn,
+                        &(
+                            Bound::Excluded(&parent_level_left_bound),
+                            Bound::Included(&parent_key_prefix),
+                        ),
+                    )?
+                    .next();
+
+                match parent {
+                    Some(Ok((parent_key, _parent_value))) => {
+                        // found parent, cache it for next keys
+                        last_parent = Some(parent_key.left_bound.to_owned().into_boxed_slice());
+
+                        // add to modified list for parent level
+                        changed_parents.push(FacetFieldIdChange {
+                            facet_value: parent_key.left_bound.to_owned().into_boxed_slice(),
+                        });
+                        self.compute_parent_group(wtxn, child_level, child.facet_value)?;
+                    }
+                    Some(Err(err)) => return Err(err.into()),
+                    None => {
+                        // no parent for that key
+                        let mut parent_it = self
+                            .db
+                            .remap_data_type::<DecodeIgnore>()
+                            .prefix_iter_mut(wtxn, &parent_level_left_bound)?;
+                        match parent_it.next() {
+                            // 1. left of the current left bound, or
+                            Some(Ok((first_key, _first_value))) => {
+                                // make sure we don't spill on the neighboring fid (level also included defensively)
+                                if first_key.field_id != self.field_id
+                                    || first_key.level != parent_level
+                                {
+                                    // max level reached, exit
+                                    drop(parent_it);
+                                    self.compute_parent_group(
+                                        wtxn,
+                                        child_level,
+                                        child.facet_value,
+                                    )?;
+                                    for child in child_it.by_ref() {
+                                        self.compute_parent_group(
+                                            wtxn,
+                                            child_level,
+                                            child.facet_value,
+                                        )?;
+                                    }
+                                    return Ok(());
+                                }
+                                // remove old left bound
+                                unsafe { parent_it.del_current()? };
+                                drop(parent_it);
+                                changed_parents.push(FacetFieldIdChange {
+                                    facet_value: child.facet_value.clone(),
+                                });
+                                self.compute_parent_group(wtxn, child_level, child.facet_value)?;
+                                // pop all elements in order to visit the new left bound
+                                let new_left_bound =
+                                    &mut changed_parents.last_mut().unwrap().facet_value;
+                                for child in child_it.by_ref() {
+                                    new_left_bound.clone_from(&child.facet_value);
+
+                                    self.compute_parent_group(
+                                        wtxn,
+                                        child_level,
+                                        child.facet_value,
+                                    )?;
+                                }
+                            }
+                            Some(Err(err)) => return Err(err.into()),
+                            // 2. max level reached, exit
+                            None => {
+                                drop(parent_it);
+                                self.compute_parent_group(wtxn, child_level, child.facet_value)?;
+                                for child in child_it.by_ref() {
+                                    self.compute_parent_group(
+                                        wtxn,
+                                        child_level,
+                                        child.facet_value,
+                                    )?;
+                                }
+                                return Ok(());
+                            }
+                        }
+                    }
+                }
+            }
+            if changed_parents.is_empty() {
+                return Ok(());
+            }
+            drop(child_it);
+            std::mem::swap(&mut changed_children, &mut changed_parents);
+            // changed_parents is now empty because changed_children was emptied by the drain
+        }
+        Ok(())
+    }
+
+    fn compute_parent_group(
+        &self,
+        wtxn: &mut RwTxn<'_>,
+        parent_level: u8,
+        parent_left_bound: Box<[u8]>,
+    ) -> Result<()> {
+        let mut range_left_bound: Vec<u8> = parent_left_bound.into();
+        if parent_level == 0 {
+            return Ok(());
+        }
+        let child_level = parent_level - 1;
+
+        let parent_key = FacetGroupKey {
+            field_id: self.field_id,
+            level: parent_level,
+            left_bound: &*range_left_bound,
+        };
+        let child_right_bound = self
+            .db
+            .remap_data_type::<DecodeIgnore>()
+            .get_greater_than(wtxn, &parent_key)?
+            .and_then(
+                |(
+                    FacetGroupKey {
+                        level: right_level,
+                        field_id: right_fid,
+                        left_bound: right_bound,
+                    },
+                    _,
+                )| {
+                    if parent_level != right_level || self.field_id != right_fid {
+                        // there was a greater key, but with a greater level or fid, so not a sibling to the parent: ignore
+                        return None;
+                    }
+                    Some(right_bound.to_owned())
+                },
+            );
+        let child_right_bound = match &child_right_bound {
+            Some(right_bound) => Bound::Excluded(FacetGroupKey {
+                left_bound: right_bound.as_slice(),
+                field_id: self.field_id,
+                level: child_level,
+            }),
+            None => Bound::Unbounded,
+        };
+
+        let child_left_key = FacetGroupKey {
+            field_id: self.field_id,
+            level: child_level,
+            left_bound: &*range_left_bound,
+        };
+        let mut child_left_bound = Bound::Included(child_left_key);
+
+        loop {
+            // do a first pass on the range to find the number of children
+            let child_count = self
+                .db
+                .remap_data_type::<DecodeIgnore>()
+                .range(wtxn, &(child_left_bound, child_right_bound))?
+                .take(self.max_group_size as usize * 2)
+                .count();
+            let mut child_it = self.db.range(wtxn, &(child_left_bound, child_right_bound))?;
+
+            // pick the right group_size depending on the number of children
+            let group_size = if child_count >= self.max_group_size as usize * 2 {
+                // more than twice the max_group_size => there will be space for at least 2 groups of max_group_size
+                self.max_group_size as usize
+            } else if child_count >= self.group_size as usize {
+                // size in [group_size, max_group_size * 2[
+                // divided by 2 it is between [group_size / 2, max_group_size[
+                // this ensures that the tree is balanced
+                child_count / 2
+            } else {
+                // take everything
+                child_count
+            };
+
+            let res: Result<_> = child_it
+                .by_ref()
+                .take(group_size)
+                // stop if we go to the next level or field id
+                .take_while(|res| match res {
+                    Ok((child_key, _)) => {
+                        child_key.field_id == self.field_id && child_key.level == child_level
+                    }
+                    Err(_) => true,
+                })
+                .try_fold(
+                    (None, FacetGroupValue { size: 0, bitmap: Default::default() }),
+                    |(bounds, mut group_value), child_res| {
+                        let (child_key, child_value) = child_res?;
+                        let bounds = match bounds {
+                            Some((left_bound, _)) => Some((left_bound, child_key.left_bound)),
+                            None => Some((child_key.left_bound, child_key.left_bound)),
+                        };
+                        // max_group_size <= u8::MAX
+                        group_value.size += 1;
+                        group_value.bitmap |= &child_value.bitmap;
+                        Ok((bounds, group_value))
+                    },
+                );
+
+            let (bounds, group_value) = res?;
+
+            let Some((group_left_bound, right_bound)) = bounds else {
+                let update_key = FacetGroupKey {
+                    field_id: self.field_id,
+                    level: parent_level,
+                    left_bound: &*range_left_bound,
+                };
+                drop(child_it);
+                if let Bound::Included(_) = child_left_bound {
+                    self.db.delete(wtxn, &update_key)?;
+                }
+
+                break;
+            };
+
+            drop(child_it);
+            let current_left_bound = group_left_bound.to_owned();
+
+            let delete_old_bound = match child_left_bound {
+                Bound::Included(bound) => {
+                    if bound.left_bound != current_left_bound {
+                        Some(range_left_bound.clone())
+                    } else {
+                        None
+                    }
+                }
+                _ => None,
+            };
+
+            range_left_bound.clear();
+            range_left_bound.extend_from_slice(right_bound);
+            let child_left_key = FacetGroupKey {
+                field_id: self.field_id,
+                level: child_level,
+                left_bound: range_left_bound.as_slice(),
+            };
+            child_left_bound = Bound::Excluded(child_left_key);
+
+            if let Some(old_bound) = delete_old_bound {
+                let update_key = FacetGroupKey {
+                    field_id: self.field_id,
+                    level: parent_level,
+                    left_bound: old_bound.as_slice(),
+                };
+                self.db.delete(wtxn, &update_key)?;
+            }
+
+            let update_key = FacetGroupKey {
+                field_id: self.field_id,
+                level: parent_level,
+                left_bound: current_left_bound.as_slice(),
+            };
+            if group_value.bitmap.is_empty() {
+                self.db.delete(wtxn, &update_key)?;
+            } else {
+                self.db.put(wtxn, &update_key, &group_value)?;
+            }
+        }
+
+        Ok(())
+    }
+
+    /// Check whether the highest level has exceeded `min_level_size` * `self.group_size`.
+    /// If it has, we must build an addition level above it.
+    /// Then check whether the highest level is under `min_level_size`.
+    /// If it has, we must remove the complete level.
+    pub(crate) fn add_or_delete_level(&self, txn: &mut RwTxn<'_>) -> Result<()> {
+        let highest_level = get_highest_level(txn, self.db, self.field_id)?;
+        let mut highest_level_prefix = vec![];
+        highest_level_prefix.extend_from_slice(&self.field_id.to_be_bytes());
+        highest_level_prefix.push(highest_level);
+
+        let size_highest_level =
+            self.db.remap_types::<Bytes, Bytes>().prefix_iter(txn, &highest_level_prefix)?.count();
+
+        if size_highest_level >= self.group_size as usize * self.min_level_size as usize {
+            self.add_level(txn, highest_level, &highest_level_prefix, size_highest_level)
+        } else if size_highest_level < self.min_level_size as usize && highest_level != 0 {
+            self.delete_level(txn, &highest_level_prefix)
+        } else {
+            Ok(())
+        }
+    }
+
+    /// Delete a level.
+    fn delete_level(&self, txn: &mut RwTxn<'_>, highest_level_prefix: &[u8]) -> Result<()> {
+        let mut to_delete = vec![];
+        let mut iter =
+            self.db.remap_types::<Bytes, Bytes>().prefix_iter(txn, highest_level_prefix)?;
+        for el in iter.by_ref() {
+            let (k, _) = el?;
+            to_delete.push(
+                FacetGroupKeyCodec::<BytesRefCodec>::bytes_decode(k)
+                    .map_err(heed::Error::Encoding)?
+                    .into_owned(),
+            );
+        }
+        drop(iter);
+        for k in to_delete {
+            self.db.delete(txn, &k.as_ref())?;
+        }
+        Ok(())
+    }
+
+    /// Build an additional level for the field id.
+    fn add_level(
+        &self,
+        txn: &mut RwTxn<'_>,
+        highest_level: u8,
+        highest_level_prefix: &[u8],
+        size_highest_level: usize,
+    ) -> Result<()> {
+        let mut groups_iter = self
+            .db
+            .remap_types::<Bytes, FacetGroupValueCodec>()
+            .prefix_iter(txn, highest_level_prefix)?;
+
+        let nbr_new_groups = size_highest_level / self.group_size as usize;
+        let nbr_leftover_elements = size_highest_level % self.group_size as usize;
+
+        let mut to_add = vec![];
+        for _ in 0..nbr_new_groups {
+            let mut first_key = None;
+            let mut values = RoaringBitmap::new();
+            for _ in 0..self.group_size {
+                let (key_bytes, value_i) = groups_iter.next().unwrap()?;
+                let key_i = FacetGroupKeyCodec::<BytesRefCodec>::bytes_decode(key_bytes)
+                    .map_err(heed::Error::Encoding)?;
+
+                if first_key.is_none() {
+                    first_key = Some(key_i);
+                }
+                values |= value_i.bitmap;
+            }
+            let key = FacetGroupKey {
+                field_id: self.field_id,
+                level: highest_level + 1,
+                left_bound: first_key.unwrap().left_bound,
+            };
+            let value = FacetGroupValue { size: self.group_size, bitmap: values };
+            to_add.push((key.into_owned(), value));
+        }
+        // now we add the rest of the level, in case its size is > group_size * min_level_size
+        // this can indeed happen if the min_level_size parameter changes between two calls to `insert`
+        if nbr_leftover_elements > 0 {
+            let mut first_key = None;
+            let mut values = RoaringBitmap::new();
+            for _ in 0..nbr_leftover_elements {
+                let (key_bytes, value_i) = groups_iter.next().unwrap()?;
+                let key_i = FacetGroupKeyCodec::<BytesRefCodec>::bytes_decode(key_bytes)
+                    .map_err(heed::Error::Encoding)?;
+
+                if first_key.is_none() {
+                    first_key = Some(key_i);
+                }
+                values |= value_i.bitmap;
+            }
+            let key = FacetGroupKey {
+                field_id: self.field_id,
+                level: highest_level + 1,
+                left_bound: first_key.unwrap().left_bound,
+            };
+            // Note: nbr_leftover_elements can be casted to a u8 since it is bounded by `max_group_size`
+            // when it is created above.
+            let value = FacetGroupValue { size: nbr_leftover_elements as u8, bitmap: values };
+            to_add.push((key.into_owned(), value));
+        }
+
+        drop(groups_iter);
+        for (key, value) in to_add {
+            self.db.put(txn, &key.as_ref(), &value)?;
+        }
+        Ok(())
+    }
+}
+
+#[derive(Debug)]
+pub struct FacetFieldIdChange {
+    pub facet_value: Box<[u8]>,
+}

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

@@ -123,7 +123,8 @@ pub fn enrich_documents_batch<R: Read + Seek>(
             }
         }
 
-        let document_id = serde_json::to_vec(&document_id).map_err(InternalError::SerdeJson)?;
+        let document_id =
+            serde_json::to_vec(&document_id).map_err(InternalError::SerdeJson).unwrap();
         external_ids.insert(count.to_be_bytes(), document_id)?;
 
         count += 1;
@@ -237,7 +238,7 @@ pub fn validate_geo_from_json(id: &DocumentId, bytes: &[u8]) -> Result<StdResult
     let debug_id = || {
         serde_json::from_slice(id.value().as_bytes()).unwrap_or_else(|_| Value::from(id.debug()))
     };
-    match serde_json::from_slice(bytes).map_err(InternalError::SerdeJson)? {
+    match serde_json::from_slice(bytes).map_err(InternalError::SerdeJson).unwrap() {
         Value::Object(mut object) => match (object.remove("lat"), object.remove("lng")) {
             (Some(lat), Some(lng)) => {
                 match (extract_finite_float_from_value(lat), extract_finite_float_from_value(lng)) {