Replace the HashMap caches by BTreeMaps

Cargo.lock

@@ -496,7 +496,7 @@ source = "git+https://github.com/meilisearch/bbqueue#cbb87cc707b5af415ef203bdaf2
 
 [[package]]
 name = "benchmarks"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "anyhow",
  "bumpalo",
@@ -689,7 +689,7 @@ dependencies = [
 
 [[package]]
 name = "build-info"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "anyhow",
  "time",
@@ -706,20 +706,6 @@ 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"
@@ -1664,7 +1650,7 @@ dependencies = [
 
 [[package]]
 name = "dump"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "anyhow",
  "big_s",
@@ -1876,7 +1862,7 @@ checksum = "486f806e73c5707928240ddc295403b1b93c96a02038563881c4a2fd84b81ac4"
 
 [[package]]
 name = "file-store"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "tempfile",
  "thiserror",
@@ -1898,7 +1884,7 @@ dependencies = [
 
 [[package]]
 name = "filter-parser"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "insta",
  "nom",
@@ -1918,7 +1904,7 @@ dependencies = [
 
 [[package]]
 name = "flatten-serde-json"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "criterion",
  "serde_json",
@@ -2057,7 +2043,7 @@ dependencies = [
 
 [[package]]
 name = "fuzzers"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "arbitrary",
  "bumpalo",
@@ -2624,15 +2610,13 @@ checksum = "206ca75c9c03ba3d4ace2460e57b189f39f43de612c2f85836e65c929701bb2d"
 
 [[package]]
 name = "index-scheduler"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "anyhow",
  "arroy 0.5.0 (registry+https://github.com/rust-lang/crates.io-index)",
  "big_s",
  "bincode",
  "bumpalo",
- "bumparaw-collections",
- "convert_case 0.6.0",
  "crossbeam-channel",
  "csv",
  "derive_builder 0.20.0",
@@ -2647,6 +2631,7 @@ dependencies = [
  "meilisearch-types",
  "memmap2",
  "page_size",
+ "raw-collections",
  "rayon",
  "roaring",
  "serde",
@@ -2662,12 +2647,12 @@ dependencies = [
 
 [[package]]
 name = "indexmap"
-version = "2.7.0"
+version = "2.2.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "62f822373a4fe84d4bb149bf54e584a7f4abec90e072ed49cda0edea5b95471f"
+checksum = "168fb715dda47215e360912c096649d23d58bf392ac62f73919e831745e40f26"
 dependencies = [
  "equivalent",
- "hashbrown 0.15.1",
+ "hashbrown 0.14.3",
  "serde",
 ]
 
@@ -2822,7 +2807,7 @@ dependencies = [
 
 [[package]]
 name = "json-depth-checker"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "criterion",
  "serde_json",
@@ -3441,7 +3426,7 @@ checksum = "490cc448043f947bae3cbee9c203358d62dbee0db12107a74be5c30ccfd09771"
 
 [[package]]
 name = "meili-snap"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "insta",
  "md5",
@@ -3450,7 +3435,7 @@ dependencies = [
 
 [[package]]
 name = "meilisearch"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "actix-cors",
  "actix-http",
@@ -3540,7 +3525,7 @@ dependencies = [
 
 [[package]]
 name = "meilisearch-auth"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "base64 0.22.1",
  "enum-iterator",
@@ -3559,12 +3544,11 @@ dependencies = [
 
 [[package]]
 name = "meilisearch-types"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "actix-web",
  "anyhow",
  "bumpalo",
- "bumparaw-collections",
  "convert_case 0.6.0",
  "csv",
  "deserr",
@@ -3577,8 +3561,8 @@ dependencies = [
  "meili-snap",
  "memmap2",
  "milli",
+ "raw-collections",
  "roaring",
- "rustc-hash 2.1.0",
  "serde",
  "serde-cs",
  "serde_json",
@@ -3592,19 +3576,16 @@ dependencies = [
 
 [[package]]
 name = "meilitool"
-version = "1.12.5"
+version = "1.12.0"
 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",
 ]
@@ -3627,7 +3608,7 @@ dependencies = [
 
 [[package]]
 name = "milli"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "allocator-api2",
  "arroy 0.5.0 (registry+https://github.com/rust-lang/crates.io-index)",
@@ -3637,7 +3618,6 @@ dependencies = [
  "bincode",
  "bstr",
  "bumpalo",
- "bumparaw-collections",
  "bytemuck",
  "byteorder",
  "candle-core",
@@ -3676,12 +3656,13 @@ dependencies = [
  "once_cell",
  "ordered-float",
  "rand",
+ "raw-collections",
  "rayon",
  "rayon-par-bridge",
  "rhai",
  "roaring",
  "rstar",
- "rustc-hash 2.1.0",
+ "rustc-hash 2.0.0",
  "serde",
  "serde_json",
  "slice-group-by",
@@ -4083,7 +4064,7 @@ checksum = "e3148f5046208a5d56bcfc03053e3ca6334e51da8dfb19b6cdc8b306fae3283e"
 
 [[package]]
 name = "permissive-json-pointer"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "big_s",
  "serde_json",
@@ -4430,7 +4411,7 @@ dependencies = [
  "bytes",
  "rand",
  "ring",
- "rustc-hash 2.1.0",
+ "rustc-hash 2.0.0",
  "rustls",
  "slab",
  "thiserror",
@@ -4506,6 +4487,19 @@ 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"
@@ -4803,9 +4797,9 @@ checksum = "08d43f7aa6b08d49f382cde6a7982047c3426db949b1424bc4b7ec9ae12c6ce2"
 
 [[package]]
 name = "rustc-hash"
-version = "2.1.0"
+version = "2.0.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c7fb8039b3032c191086b10f11f319a6e99e1e82889c5cc6046f515c9db1d497"
+checksum = "583034fd73374156e66797ed8e5b0d5690409c9226b22d87cb7f19821c05d152"
 
 [[package]]
 name = "rustc_version"
@@ -4974,9 +4968,9 @@ dependencies = [
 
 [[package]]
 name = "serde_json"
-version = "1.0.133"
+version = "1.0.132"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c7fceb2473b9166b2294ef05efcb65a3db80803f0b03ef86a5fc88a2b85ee377"
+checksum = "d726bfaff4b320266d395898905d0eba0345aae23b54aee3a737e260fd46db03"
 dependencies = [
  "indexmap",
  "itoa",
@@ -6486,7 +6480,7 @@ dependencies = [
 
 [[package]]
 name = "xtask"
-version = "1.12.5"
+version = "1.12.0"
 dependencies = [
  "anyhow",
  "build-info",

Cargo.toml

@@ -22,7 +22,7 @@ members = [
 ]
 
 [workspace.package]
-version = "1.12.5"
+version = "1.12.0"
 authors = [
     "Quentin de Quelen <quentin@dequelen.me>",
     "Clément Renault <clement@meilisearch.com>",

crates/benchmarks/benches/indexing.rs

@@ -8,7 +8,6 @@ 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;
@@ -152,7 +151,7 @@ fn indexing_songs_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -167,7 +166,7 @@ fn indexing_songs_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -219,7 +218,7 @@ fn reindexing_songs_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -234,7 +233,7 @@ fn reindexing_songs_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -264,7 +263,7 @@ fn reindexing_songs_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -279,7 +278,7 @@ fn reindexing_songs_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -333,7 +332,7 @@ fn deleting_songs_in_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -348,7 +347,7 @@ fn deleting_songs_in_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -410,7 +409,7 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -425,7 +424,7 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -455,7 +454,7 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -470,7 +469,7 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -496,7 +495,7 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -511,7 +510,7 @@ fn indexing_songs_in_three_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -564,7 +563,7 @@ fn indexing_songs_without_faceted_numbers(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -579,7 +578,7 @@ fn indexing_songs_without_faceted_numbers(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -631,7 +630,7 @@ fn indexing_songs_without_faceted_fields(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -646,7 +645,7 @@ fn indexing_songs_without_faceted_fields(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -698,7 +697,7 @@ fn indexing_wiki(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -713,7 +712,7 @@ fn indexing_wiki(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -764,7 +763,7 @@ fn reindexing_wiki(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -779,7 +778,7 @@ fn reindexing_wiki(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -809,7 +808,7 @@ fn reindexing_wiki(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -824,7 +823,7 @@ fn reindexing_wiki(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -877,7 +876,7 @@ fn deleting_wiki_in_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -892,7 +891,7 @@ fn deleting_wiki_in_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -954,7 +953,7 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -969,7 +968,7 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1000,7 +999,7 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -1015,7 +1014,7 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1042,7 +1041,7 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -1057,7 +1056,7 @@ fn indexing_wiki_in_three_batches(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1109,7 +1108,7 @@ fn indexing_movies_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -1124,7 +1123,7 @@ fn indexing_movies_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1175,7 +1174,7 @@ fn reindexing_movies_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -1190,7 +1189,7 @@ fn reindexing_movies_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1220,7 +1219,7 @@ fn reindexing_movies_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -1235,7 +1234,7 @@ fn reindexing_movies_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1288,7 +1287,7 @@ fn deleting_movies_in_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -1303,7 +1302,7 @@ fn deleting_movies_in_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1351,7 +1350,7 @@ fn delete_documents_from_ids(index: Index, document_ids_to_delete: Vec<RoaringBi
             &document_changes,
             EmbeddingConfigs::default(),
             &|| false,
-            &Progress::default(),
+            &|_| (),
         )
         .unwrap();
 
@@ -1401,7 +1400,7 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -1416,7 +1415,7 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1446,7 +1445,7 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -1461,7 +1460,7 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1487,7 +1486,7 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -1502,7 +1501,7 @@ fn indexing_movies_in_three_batches(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1577,7 +1576,7 @@ fn indexing_nested_movies_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -1592,7 +1591,7 @@ fn indexing_nested_movies_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1668,7 +1667,7 @@ fn deleting_nested_movies_in_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -1683,7 +1682,7 @@ fn deleting_nested_movies_in_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1751,7 +1750,7 @@ fn indexing_nested_movies_without_faceted_fields(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -1766,7 +1765,7 @@ fn indexing_nested_movies_without_faceted_fields(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1818,7 +1817,7 @@ fn indexing_geo(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -1833,7 +1832,7 @@ fn indexing_geo(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1884,7 +1883,7 @@ fn reindexing_geo(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -1899,7 +1898,7 @@ fn reindexing_geo(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1929,7 +1928,7 @@ fn reindexing_geo(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -1944,7 +1943,7 @@ fn reindexing_geo(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 
@@ -1997,7 +1996,7 @@ fn deleting_geo_in_batches_default(c: &mut Criterion) {
                         None,
                         &mut new_fields_ids_map,
                         &|| false,
-                        Progress::default(),
+                        &|_progress| (),
                     )
                     .unwrap();
 
@@ -2012,7 +2011,7 @@ fn deleting_geo_in_batches_default(c: &mut Criterion) {
                     &document_changes,
                     EmbeddingConfigs::default(),
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
                 .unwrap();
 

crates/benchmarks/benches/utils.rs

@@ -10,7 +10,6 @@ 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;
@@ -111,7 +110,7 @@ pub fn base_setup(conf: &Conf) -> Index {
             None,
             &mut new_fields_ids_map,
             &|| false,
-            Progress::default(),
+            &|_progress| (),
         )
         .unwrap();
 
@@ -126,7 +125,7 @@ pub fn base_setup(conf: &Conf) -> Index {
         &document_changes,
         EmbeddingConfigs::default(),
         &|| false,
-        &Progress::default(),
+        &|_| (),
     )
     .unwrap();
 

crates/file-store/src/lib.rs

@@ -136,14 +136,6 @@ 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

@@ -10,7 +10,6 @@ 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;
@@ -129,7 +128,7 @@ fn main() {
                                     None,
                                     &mut new_fields_ids_map,
                                     &|| false,
-                                    Progress::default(),
+                                    &|_progress| (),
                                 )
                                 .unwrap();
 
@@ -144,7 +143,7 @@ fn main() {
                                 &document_changes,
                                 embedders,
                                 &|| false,
-                                &Progress::default(),
+                                &|_| (),
                             )
                             .unwrap();
 

crates/index-scheduler/Cargo.toml

@@ -13,9 +13,6 @@ 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" }
@@ -24,8 +21,8 @@ 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 = "0.10.7", features = ["serde"] }
 serde = { version = "1.0.204", features = ["derive"] }
@@ -33,6 +30,7 @@ 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",
@@ -42,6 +40,7 @@ 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"

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 400 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 512 bytes."
     )]
     InvalidIndexUid { index_uid: String },
     #[error("Task `{0}` not found.")]
@@ -122,11 +122,8 @@ pub enum Error {
     Dump(#[from] dump::Error),
     #[error(transparent)]
     Heed(#[from] heed::Error),
-    #[error("{}", match .index_uid {
+    #[error(transparent)]
-        Some(uid) if !uid.is_empty() => format!("Index `{}`: {error}", uid),
+    Milli(#[from] milli::Error),
-        _ => format!("{error}")
-    })]
-    Milli { error: milli::Error, index_uid: Option<String> },
     #[error("An unexpected crash occurred when processing the task.")]
     ProcessBatchPanicked,
     #[error(transparent)]
@@ -193,7 +190,7 @@ impl Error {
             | Error::AbortedTask
             | Error::Dump(_)
             | Error::Heed(_)
-            | Error::Milli { .. }
+            | Error::Milli(_)
             | Error::ProcessBatchPanicked
             | Error::FileStore(_)
             | Error::IoError(_)
@@ -212,20 +209,6 @@ 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 {
@@ -253,7 +236,7 @@ impl ErrorCode for Error {
             // TODO: not sure of the Code to use
             Error::NoSpaceLeftInTaskQueue => Code::NoSpaceLeftOnDevice,
             Error::Dump(e) => e.error_code(),
-            Error::Milli { error, .. } => error.error_code(),
+            Error::Milli(e) => e.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

@@ -3,13 +3,14 @@ use std::path::Path;
 use std::time::Duration;
 
 use meilisearch_types::heed::{EnvClosingEvent, EnvFlags, EnvOpenOptions};
-use meilisearch_types::milli::{Index, Result};
+use meilisearch_types::milli::Index;
 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.

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

@@ -5,7 +5,6 @@ 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};
@@ -122,7 +121,7 @@ impl IndexStats {
     /// # Parameters
     ///
     /// - rtxn: a RO transaction for the index, obtained from `Index::read_txn()`.
-    pub fn new(index: &Index, rtxn: &RoTxn) -> milli::Result<Self> {
+    pub fn new(index: &Index, rtxn: &RoTxn) -> Result<Self> {
         Ok(IndexStats {
             number_of_documents: index.number_of_documents(rtxn)?,
             database_size: index.on_disk_size()?,
@@ -184,18 +183,13 @@ 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
+                let index = self.index_map.write().unwrap().create(
-                    .index_map
+                    &uuid,
-                    .write()
+                    &index_path,
-                    .unwrap()
+                    date,
-                    .create(
+                    self.enable_mdb_writemap,
-                        &uuid,
+                    self.index_base_map_size,
-                        &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()?;
 
@@ -363,9 +357,7 @@ impl IndexMapper {
                     };
                     let index_path = self.base_path.join(uuid.to_string());
                     // take the lock to reopen the environment.
-                    reopen
+                    reopen.reopen(&mut self.index_map.write().unwrap(), &index_path)?;
-                        .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())),
@@ -380,15 +372,13 @@ impl IndexMapper {
                         Missing => {
                             let index_path = self.base_path.join(uuid.to_string());
 
-                            break index_map
+                            break index_map.create(
-                                .create(
+                                &uuid,
-                                    &uuid,
+                                &index_path,
-                                    &index_path,
+                                None,
-                                    None,
+                                self.enable_mdb_writemap,
-                                    self.enable_mdb_writemap,
+                                self.index_base_map_size,
-                                    self.index_base_map_size,
+                            )?;
-                                )
-                                .map_err(|e| Error::from_milli(e, Some(uuid.to_string())))?;
                         }
                         Available(index) => break index,
                         Closing(_) => {
@@ -470,7 +460,6 @@ 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, progress: _ } = batch;
+    let Batch { uid, details, stats, started_at, finished_at } = batch;
     if let Some(finished_at) = finished_at {
         assert!(finished_at > started_at);
     }

crates/index-scheduler/src/lib.rs

@@ -26,7 +26,6 @@ mod index_mapper;
 #[cfg(test)]
 mod insta_snapshot;
 mod lru;
-mod processing;
 mod utils;
 pub mod uuid_codec;
 
@@ -55,13 +54,14 @@ 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};
+use meilisearch_types::tasks::{Kind, KindWithContent, Status, Task, TaskProgress};
-use processing::ProcessingTasks;
 use rayon::current_num_threads;
 use rayon::prelude::{IntoParallelIterator, ParallelIterator};
 use roaring::RoaringBitmap;
@@ -72,8 +72,7 @@ use utils::{filter_out_references_to_newer_tasks, keep_ids_within_datetimes, map
 use uuid::Uuid;
 
 use crate::index_mapper::IndexMapper;
-use crate::processing::{AtomicTaskStep, BatchProgress};
+use crate::utils::{check_index_swap_validity, clamp_to_page_size, ProcessingBatch};
-use crate::utils::{check_index_swap_validity, clamp_to_page_size};
 
 pub(crate) type BEI128 = I128<BE>;
 
@@ -164,6 +163,48 @@ 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>);
 
@@ -772,7 +813,7 @@ 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)?;
                 }
@@ -786,13 +827,13 @@ 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;
         }
@@ -841,7 +882,7 @@ 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
@@ -1049,7 +1090,7 @@ 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
@@ -1565,8 +1606,7 @@ impl IndexScheduler {
 
         // We reset the must_stop flag to be sure that we don't stop processing tasks
         self.must_stop_processing.reset();
-        let progress = self
+        self.processing_tasks
-            .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
@@ -1579,12 +1619,11 @@ 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, progress)
+                        cloned_index_scheduler.process_batch(batch, processing_batch)
                     })
                     .unwrap();
                 handle.join().unwrap_or(Err(Error::ProcessBatchPanicked))
@@ -1597,7 +1636,6 @@ 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();
@@ -1607,15 +1645,12 @@ 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);
@@ -1643,10 +1678,9 @@ 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 {
+            Err(Error::Milli(milli::Error::InternalError(
-                error: milli::Error::InternalError(milli::InternalError::AbortedIndexation),
+                milli::InternalError::AbortedIndexation,
-                ..
+            )))
-            })
             | Err(Error::AbortedTask) => {
                 #[cfg(test)]
                 self.breakpoint(Breakpoint::AbortedIndexation);
@@ -1665,10 +1699,9 @@ impl IndexScheduler {
             // 2. close the associated environment
             // 3. resize it
             // 4. re-schedule tasks
-            Err(Error::Milli {
+            Err(Error::Milli(milli::Error::UserError(
-                error: milli::Error::UserError(milli::UserError::MaxDatabaseSizeReached),
+                milli::UserError::MaxDatabaseSizeReached,
-                ..
+            ))) if index_uid.is_some() => {
-            }) 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
@@ -1683,12 +1716,8 @@ 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)))?
@@ -1914,7 +1943,6 @@ 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
@@ -1925,12 +1953,8 @@ impl IndexScheduler {
                      config: milli::vector::EmbeddingConfig { embedder_options, prompt, quantized },
                      ..
                  }| {
-                    let prompt = Arc::new(
+                    let prompt =
-                        prompt
+                        Arc::new(prompt.try_into().map_err(meilisearch_types::milli::Error::from)?);
-                            .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();
@@ -1946,9 +1970,7 @@ impl IndexScheduler {
                     let embedder = Arc::new(
                         Embedder::new(embedder_options.clone())
                             .map_err(meilisearch_types::milli::vector::Error::from)
-                            .map_err(|err| {
+                            .map_err(meilisearch_types::milli::Error::from)?,
-                                Error::from_milli(err.into(), Some(index_uid.clone()))
-                            })?,
                     );
                     {
                         let mut embedders = self.embedders.write().unwrap();
@@ -2016,19 +2038,14 @@ 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, file) = self.index_scheduler.create_update_file(false)?;
+                let (uuid, mut file) = self.index_scheduler.create_update_file(false)?;
-                let mut writer = io::BufWriter::new(file);
+                let mut builder = DocumentsBatchBuilder::new(&mut file);
                 for doc in content_file {
-                    let doc = doc?;
+                    builder.append_json_object(&doc?)?;
-                    serde_json::to_writer(&mut writer, &doc).map_err(|e| {
-                        Error::from_milli(milli::InternalError::SerdeJson(e).into(), None)
-                    })?;
                 }
-                let file = writer.into_inner().map_err(|e| e.into_error())?;
+                builder.into_inner()?;
                 file.persist()?;
 
                 Some(uuid)
@@ -2036,12 +2053,6 @@ 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,
         };
 
@@ -4308,35 +4319,10 @@ mod tests {
         let proc = index_scheduler.processing_tasks.read().unwrap().clone();
 
         let query = Query { statuses: Some(vec![Status::Processing]), ..Default::default() };
-        let (mut batches, _) = index_scheduler
+        let (batches, _) = index_scheduler
-            .get_batches_from_authorized_indexes(query.clone(), &AuthFilter::default())
+            .get_batch_ids_from_authorized_indexes(&rtxn, &proc, &query, &AuthFilter::default())
             .unwrap();
-        assert_eq!(batches.len(), 1);
+        snapshot!(snapshot_bitmap(&batches), @"[0,]"); // only the processing batch in the first tick
-        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
@@ -6160,7 +6146,7 @@ mod tests {
             insta::assert_json_snapshot!(simple_hf_config.embedder_options);
             let simple_hf_name = name.clone();
 
-            let configs = index_scheduler.embedders("doggos".to_string(), configs).unwrap();
+            let configs = index_scheduler.embedders(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

@@ -1,316 +0,0 @@
-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":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"beavero":1}}, }
+{uid: 1, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"beavero":2}}, }
 ----------------------------------------------------------------------
 ### 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":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"beavero":1}}, }
+{uid: 1, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"beavero":2}}, }
 ----------------------------------------------------------------------
 ### 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":{"processing":1},"types":{"dumpCreation":1},"indexUids":{}}, }
+{uid: 0, details: {"dumpUid":null}, stats: {"totalNbTasks":1,"status":{"enqueued":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":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"catto":1}}, }
+{uid: 0, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"catto":2}}, }
 ----------------------------------------------------------------------
 ### 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":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"catto":1}}, }
+{uid: 0, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"catto":2}}, }
 ----------------------------------------------------------------------
 ### 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":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"catto":1}}, }
+{uid: 0, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"catto":2}}, }
 ----------------------------------------------------------------------
 ### 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":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
+{uid: 0, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"doggos":2}}, }
 ----------------------------------------------------------------------
 ### 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":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
+{uid: 0, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"doggos":2}}, }
 ----------------------------------------------------------------------
 ### 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: "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) }}
+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: "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") }}
+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") }}
 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":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
+{uid: 0, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"doggos":2}}, }
 ----------------------------------------------------------------------
 ### 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":1,"indexedDocuments":null}, stats: {"totalNbTasks":1,"status":{"processing":1},"types":{"documentAdditionOrUpdate":1},"indexUids":{"doggos":1}}, }
+{uid: 0, details: {"receivedDocuments":2,"indexedDocuments":null}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"documentAdditionOrUpdate":2},"indexUids":{"doggos":2}}, }
 ----------------------------------------------------------------------
 ### 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":1,"status":{"processing":1},"types":{"indexCreation":1},"indexUids":{"index_a":1}}, }
+{uid: 0, details: {"primaryKey":"id"}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"indexCreation":2},"indexUids":{"index_a":2}}, }
 ----------------------------------------------------------------------
 ### 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":1,"status":{"processing":1},"types":{"indexCreation":1},"indexUids":{"index_a":1}}, }
+{uid: 0, details: {"primaryKey":"id"}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"indexCreation":2},"indexUids":{"index_a":2}}, }
 ----------------------------------------------------------------------
 ### 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":1,"status":{"processing":1},"types":{"indexCreation":1},"indexUids":{"index_a":1}}, }
+{uid: 0, details: {"primaryKey":"id"}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"indexCreation":2},"indexUids":{"index_a":2}}, }
 ----------------------------------------------------------------------
 ### 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":1,"status":{"processing":1},"types":{"indexCreation":1},"indexUids":{"doggo":1}}, }
+{uid: 1, details: {"primaryKey":"sheep"}, stats: {"totalNbTasks":2,"status":{"enqueued":2},"types":{"indexCreation":2},"indexUids":{"doggo":2}}, }
 ----------------------------------------------------------------------
 ### 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":{"processing":1},"types":{"taskDeletion":1},"indexUids":{}}, }
+{uid: 0, details: {"matchedTasks":2,"deletedTasks":null,"originalFilter":"test_query"}, stats: {"totalNbTasks":1,"status":{"enqueued":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(Status::Processing).or_default() += 1;
+            *self.stats.status.entry(task.status).or_default() += 1;
 
             self.kinds.insert(task.kind.as_kind());
             *self.stats.types.entry(task.kind.as_kind()).or_default() += 1;
@@ -134,7 +134,6 @@ 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,
@@ -188,7 +187,6 @@ impl IndexScheduler {
             &batch.uid,
             &Batch {
                 uid: batch.uid,
-                progress: None,
                 details: batch.details,
                 stats: batch.stats,
                 started_at: batch.started_at,
@@ -275,9 +273,7 @@ impl IndexScheduler {
             .into_iter()
             .map(|batch_id| {
                 if Some(batch_id) == processing.batch.as_ref().map(|batch| batch.uid) {
-                    let mut batch = processing.batch.as_ref().unwrap().to_batch();
+                    Ok(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))
@@ -291,10 +287,7 @@ impl IndexScheduler {
 
         debug_assert!(old_task != *task);
         debug_assert_eq!(old_task.uid, task.uid);
-        debug_assert!(
+        debug_assert!(old_task.batch_uid.is_none() && task.batch_uid.is_some());
-            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-types/Cargo.toml

@@ -24,9 +24,8 @@ flate2 = "1.0.30"
 fst = "0.4.7"
 memmap2 = "0.9.4"
 milli = { path = "../milli" }
-bumparaw-collections = "0.1.2"
+raw-collections = { git = "https://github.com/meilisearch/raw-collections.git", version = "0.1.0" }
 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};
+use crate::{
-use crate::task_view::DetailsView;
+    batches::{Batch, BatchId, BatchStats},
-use crate::tasks::serialize_duration;
+    task_view::DetailsView,
+    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)]
@@ -25,7 +25,6 @@ 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

@@ -1,11 +1,12 @@
 use std::collections::BTreeMap;
 
-use milli::progress::ProgressView;
 use serde::{Deserialize, Serialize};
 use time::OffsetDateTime;
 
-use crate::task_view::DetailsView;
+use crate::{
-use crate::tasks::{Kind, Status};
+    task_view::DetailsView,
+    tasks::{Kind, Status},
+};
 
 pub type BatchId = u32;
 
@@ -14,8 +15,6 @@ 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

@@ -4,11 +4,10 @@ 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 rustc_hash::FxBuildHasher;
+use raw_collections::RawMap;
 use serde::de::{SeqAccess, Visitor};
 use serde::{Deserialize, Deserializer};
 use serde_json::error::Category;
@@ -221,7 +220,7 @@ pub fn read_json(input: &File, output: impl io::Write) -> Result<u64> {
     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_and_hasher(obj, FxBuildHasher, &doc_alloc)?;
+        let map = RawMap::from_raw_value(obj, &doc_alloc)?;
         to_writer(&mut out, &map)
     });
     let count = match res {
@@ -251,25 +250,26 @@ pub fn read_json(input: &File, output: impl io::Write) -> Result<u64> {
     }
 }
 
-/// Reads NDJSON from file and checks it.
+/// Reads NDJSON from file and write it in NDJSON in a file checking it along the way.
-pub fn read_ndjson(input: &File) -> Result<u64> {
+pub fn read_ndjson(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 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();
-        match result {
+        count += 1;
-            Ok(raw) => {
+        result
+            .and_then(|raw: &RawValue| {
                 // try to deserialize as a map
-                RawMap::from_raw_value_and_hasher(raw, FxBuildHasher, &bump)
+                let map = RawMap::from_raw_value(raw, &bump)?;
-                    .map_err(|e| DocumentFormatError::from((PayloadType::Ndjson, e)))?;
+                to_writer(&mut output, &map)
-                count += 1;
+            })
-            }
+            .map_err(|e| DocumentFormatError::from((PayloadType::Ndjson, e)))?;
-            Err(e) => return Err(DocumentFormatError::from((PayloadType::Ndjson, e))),
-        }
     }
 
     Ok(count)

crates/meilisearch-types/src/error.rs

@@ -279,7 +279,6 @@ 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 ;
@@ -550,7 +549,7 @@ 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 511 bytes."
+            and can not be more than 512 bytes."
         )
     }
 }

crates/meilisearch-types/src/tasks.rs

@@ -4,6 +4,7 @@ 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;
@@ -40,6 +41,62 @@ 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/src/error.rs

@@ -4,7 +4,6 @@ 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;
@@ -63,11 +62,8 @@ pub enum MeilisearchHttpError {
     HeedError(#[from] meilisearch_types::heed::Error),
     #[error(transparent)]
     IndexScheduler(#[from] index_scheduler::Error),
-    #[error("{}", match .index_name {
+    #[error(transparent)]
-        Some(name) if !name.is_empty() => format!("Index `{}`: {error}", name),
+    Milli(#[from] meilisearch_types::milli::Error),
-        _ => format!("{error}")
-    })]
-    Milli { error: milli::Error, index_name: Option<String> },
     #[error(transparent)]
     Payload(#[from] PayloadError),
     #[error(transparent)]
@@ -80,12 +76,6 @@ 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 {
@@ -105,7 +95,7 @@ impl ErrorCode for MeilisearchHttpError {
             MeilisearchHttpError::SerdeJson(_) => Code::Internal,
             MeilisearchHttpError::HeedError(_) => Code::Internal,
             MeilisearchHttpError::IndexScheduler(e) => e.error_code(),
-            MeilisearchHttpError::Milli { error, .. } => error.error_code(),
+            MeilisearchHttpError::Milli(e) => e.error_code(),
             MeilisearchHttpError::Payload(e) => e.error_code(),
             MeilisearchHttpError::FileStore(_) => Code::Internal,
             MeilisearchHttpError::DocumentFormat(e) => e.error_code(),

crates/meilisearch/src/lib.rs

@@ -395,7 +395,6 @@ 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));
@@ -433,7 +432,7 @@ fn import_dump(
         let reader = DocumentsBatchReader::from_reader(reader)?;
 
         let embedder_configs = index.embedding_configs(&wtxn)?;
-        let embedders = index_scheduler.embedders(uid, embedder_configs)?;
+        let embedders = index_scheduler.embedders(embedder_configs)?;
 
         let builder = milli::update::IndexDocuments::new(
             &mut wtxn,

crates/meilisearch/src/main.rs

@@ -129,11 +129,6 @@ 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/option.rs

@@ -654,8 +654,9 @@ impl Opt {
 
 #[derive(Debug, Default, Clone, Parser, Deserialize)]
 pub struct IndexerOpts {
-    /// Sets the maximum amount of RAM Meilisearch can use when indexing. By default, Meilisearch
+    /// Specifies the maximum resident memory that Meilisearch can use for indexing.
-    /// uses no more than two thirds of available memory.
+    /// By default, Meilisearch limits the RAM usage to 5% of the total available memory.
+    /// Note that the underlying store utilizes memory-mapping and makes use of the rest.
     #[clap(long, env = MEILI_MAX_INDEXING_MEMORY, default_value_t)]
     #[serde(default)]
     pub max_indexing_memory: MaxMemory,
@@ -714,7 +715,7 @@ impl TryFrom<&IndexerOpts> for IndexerConfig {
     }
 }
 
-/// A type used to detect the max memory available and use 2/3 of it.
+/// A type used to detect the max resident memory available and use 5% of it.
 #[derive(Debug, Clone, Copy, Deserialize, Serialize)]
 pub struct MaxMemory(Option<Byte>);
 
@@ -728,7 +729,7 @@ impl FromStr for MaxMemory {
 
 impl Default for MaxMemory {
     fn default() -> MaxMemory {
-        MaxMemory(total_memory_bytes().map(|bytes| bytes * 2 / 3).map(Byte::from_u64))
+        MaxMemory(total_memory_bytes().map(|bytes| bytes * 5 / 100).map(Byte::from_u64))
     }
 }
 

crates/meilisearch/src/routes/batches.rs

@@ -1,18 +1,18 @@
-use actix_web::web::{self, Data};
+use actix_web::{
-use actix_web::HttpResponse;
+    web::{self, Data},
+    HttpResponse,
+};
 use deserr::actix_web::AwebQueryParameter;
 use index_scheduler::{IndexScheduler, Query};
-use meilisearch_types::batch_view::BatchView;
+use meilisearch_types::{
-use meilisearch_types::batches::BatchId;
+    batch_view::BatchView, batches::BatchId, deserr::DeserrQueryParamError, error::ResponseError,
-use meilisearch_types::deserr::DeserrQueryParamError;
+    keys::actions,
-use meilisearch_types::error::ResponseError;
+};
-use meilisearch_types::keys::actions;
 use serde::Serialize;
 
-use super::tasks::TasksFilterQuery;
+use crate::extractors::{authentication::GuardedData, sequential_extractor::SeqHandler};
-use super::ActionPolicy;
+
-use crate::extractors::authentication::GuardedData;
+use super::{tasks::TasksFilterQuery, ActionPolicy};
-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, Seek as _};
+use std::io::ErrorKind;
 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>,
-    body: Payload,
+    mut body: Payload,
     method: IndexDocumentsMethod,
     task_id: Option<TaskId>,
     dry_run: bool,
@@ -609,60 +609,54 @@ 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 documents_count = tokio::task::spawn_blocking(move || {
+    let temp_file = match tempfile() {
-                let documents_count = file.as_ref().map_or(Ok(0), |ntf| {
+        Ok(file) => file,
-                    read_ndjson(ntf.as_file()).map_err(MeilisearchHttpError::DocumentFormat)
+        Err(e) => return Err(MeilisearchHttpError::Payload(ReceivePayload(Box::new(e)))),
-                })?;
-
-                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.
@@ -709,39 +703,6 @@ 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

@@ -185,8 +185,7 @@ pub async fn search(
 
     let index = index_scheduler.index(&index_uid)?;
     let features = index_scheduler.features();
-    let search_kind =
+    let search_kind = search_kind(&search_query, &index_scheduler, &index, features)?;
-        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::{Error, IndexScheduler};
+use index_scheduler::IndexScheduler;
 use meilisearch_types::deserr::query_params::Param;
 use meilisearch_types::deserr::{immutable_field_error, DeserrJsonError, DeserrQueryParamError};
 use meilisearch_types::error::deserr_codes::*;
@@ -107,10 +107,7 @@ pub async fn list_indexes(
             if !filters.is_index_authorized(uid) {
                 return Ok(None);
             }
-            Ok(Some(
+            Ok(Some(IndexView::new(uid.to_string(), index)?))
-                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

@@ -243,19 +243,11 @@ pub async fn search_with_url_query(
     let index = index_scheduler.index(&index_uid)?;
     let features = index_scheduler.features();
 
-    let search_kind =
+    let search_kind = search_kind(&query, index_scheduler.get_ref(), &index, features)?;
-        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(
+        perform_search(&index, query, search_kind, retrieve_vector, index_scheduler.features())
-            index_uid.to_string(),
-            &index,
-            query,
-            search_kind,
-            retrieve_vector,
-            index_scheduler.features(),
-        )
     })
     .await;
     permit.drop().await;
@@ -295,20 +287,12 @@ pub async fn search_with_post(
 
     let features = index_scheduler.features();
 
-    let search_kind =
+    let search_kind = search_kind(&query, index_scheduler.get_ref(), &index, features)?;
-        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(
+        perform_search(&index, query, search_kind, retrieve_vectors, index_scheduler.features())
-            index_uid.to_string(),
-            &index,
-            query,
-            search_kind,
-            retrieve_vectors,
-            index_scheduler.features(),
-        )
     })
     .await;
     permit.drop().await;
@@ -330,7 +314,6 @@ 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> {
@@ -349,7 +332,7 @@ 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_uid, index, embedder, v.map(|v| v.len()))
+            SearchKind::semantic(index_scheduler, index, embedder, v.map(|v| v.len()))
         }
         // hybrid.semantic_ratio == 0.0 => keyword
         (_, Some(HybridQuery { semantic_ratio, embedder: _ }), _) if **semantic_ratio == 0.0 => {
@@ -357,14 +340,13 @@ pub fn search_kind(
         }
         // no query, hybrid, vector => semantic
         (None, Some(HybridQuery { semantic_ratio: _, embedder }), Some(v)) => {
-            SearchKind::semantic(index_scheduler, index_uid, index, embedder, Some(v.len()))
+            SearchKind::semantic(index_scheduler, 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/similar.rs

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

crates/meilisearch/src/routes/multi_search.rs

@@ -125,28 +125,14 @@ pub async fn multi_search_with_post(
                         })
                         .with_index(query_index)?;
 
-                    let index_uid_str = index_uid.to_string();
+                    let search_kind =
-
+                        search_kind(&query, index_scheduler.get_ref(), &index, features)
-                    let search_kind = search_kind(
+                            .with_index(query_index)?;
-                        &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(
+                        perform_search(&index, query, search_kind, retrieve_vector, features)
-                            index_uid_str.clone(),
-                            &index,
-                            query,
-                            search_kind,
-                            retrieve_vector,
-                            features,
-                        )
                     })
                     .await
                     .with_index(query_index)?;

crates/meilisearch/src/search/federated.rs

@@ -560,8 +560,7 @@ 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 =
+                let search_kind = search_kind(&query, index_scheduler, &index, features)?;
-                    search_kind(&query, index_scheduler, index_uid.to_string(), &index, features)?;
 
                 let canonicalization_kind = match (&search_kind, &query.q) {
                     (SearchKind::SemanticOnly { .. }, _) => {
@@ -637,8 +636,7 @@ pub fn perform_federated_search(
                 search.offset(0);
                 search.limit(required_hit_count);
 
-                let (result, _semantic_hit_count) =
+                let (result, _semantic_hit_count) = super::search_from_kind(search_kind, search)?;
-                    super::search_from_kind(index_uid.to_string(), search_kind, search)?;
                 let format = AttributesFormat {
                     attributes_to_retrieve: query.attributes_to_retrieve,
                     retrieve_vectors,
@@ -672,10 +670,7 @@ pub fn perform_federated_search(
 
                 let formatter_builder = HitMaker::formatter_builder(matching_words, tokenizer);
 
-                let hit_maker =
+                let hit_maker = HitMaker::new(&index, &rtxn, format, formatter_builder)?;
-                    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

@@ -19,9 +19,7 @@ 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::{
+use meilisearch_types::milli::{FacetValueHit, OrderBy, SearchForFacetValues, TimeBudget};
-    FacetValueHit, InternalError, OrderBy, SearchForFacetValues, TimeBudget,
-};
 use meilisearch_types::settings::DEFAULT_PAGINATION_MAX_TOTAL_HITS;
 use meilisearch_types::{milli, Document};
 use milli::tokenizer::{Language, TokenizerBuilder};
@@ -283,38 +281,35 @@ 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_uid, index, embedder_name, vector_len)?;
+            Self::embedder(index_scheduler, 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_uid, index, embedder_name, vector_len)?;
+            Self::embedder(index_scheduler, 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(index_uid, embedder_configs)?;
+        let embedders = index_scheduler.embedders(embedder_configs)?;
 
         let (embedder, _, quantized) = embedders
             .get(embedder_name)
@@ -895,7 +890,6 @@ fn prepare_search<'t>(
 }
 
 pub fn perform_search(
-    index_uid: String,
     index: &Index,
     query: SearchQuery,
     search_kind: SearchKind,
@@ -922,7 +916,7 @@ pub fn perform_search(
             used_negative_operator,
         },
         semantic_hit_count,
-    ) = search_from_kind(index_uid, search_kind, search)?;
+    ) = search_from_kind(search_kind, search)?;
 
     let SearchQuery {
         q,
@@ -1075,27 +1069,17 @@ 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 => {
+        SearchKind::KeywordOnly => (search.execute()?, None),
-            let results = search
-                .execute()
-                .map_err(|e| MeilisearchHttpError::from_milli(e, Some(index_uid.to_string())))?;
-            (results, None)
-        }
         SearchKind::SemanticOnly { .. } => {
-            let results = search
+            let results = search.execute()?;
-                .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
+        SearchKind::Hybrid { semantic_ratio, .. } => search.execute_hybrid(*semantic_ratio)?,
-            .execute_hybrid(*semantic_ratio)
-            .map_err(|e| MeilisearchHttpError::from_milli(e, Some(index_uid)))?,
     };
     Ok((milli_result, semantic_hit_count))
 }
@@ -1197,7 +1181,7 @@ impl<'a> HitMaker<'a> {
         rtxn: &'a RoTxn<'a>,
         format: AttributesFormat,
         mut formatter_builder: MatcherBuilder<'a>,
-    ) -> milli::Result<Self> {
+    ) -> Result<Self, MeilisearchHttpError> {
         formatter_builder.crop_marker(format.crop_marker);
         formatter_builder.highlight_prefix(format.highlight_pre_tag);
         formatter_builder.highlight_suffix(format.highlight_post_tag);
@@ -1292,7 +1276,11 @@ impl<'a> HitMaker<'a> {
         })
     }
 
-    pub fn make_hit(&self, id: u32, score: &[ScoreDetails]) -> milli::Result<SearchHit> {
+    pub fn make_hit(
+        &self,
+        id: u32,
+        score: &[ScoreDetails],
+    ) -> Result<SearchHit, MeilisearchHttpError> {
         let (_, obkv) =
             self.index.iter_documents(self.rtxn, std::iter::once(id))?.next().unwrap()?;
 
@@ -1335,10 +1323,7 @@ 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(
+                vectors.insert(name, serde_json::to_value(embeddings)?);
-                    name,
-                    serde_json::to_value(embeddings).map_err(InternalError::SerdeJson)?,
-                );
             }
             document.insert("_vectors".into(), vectors.into());
         }
@@ -1384,7 +1369,7 @@ fn make_hits<'a>(
     format: AttributesFormat,
     matching_words: milli::MatchingWords,
     documents_ids_scores: impl Iterator<Item = (u32, &'a Vec<ScoreDetails>)> + 'a,
-) -> milli::Result<Vec<SearchHit>> {
+) -> Result<Vec<SearchHit>, MeilisearchHttpError> {
     let mut documents = Vec::new();
 
     let dictionary = index.dictionary(rtxn)?;
@@ -1422,13 +1407,6 @@ 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.
@@ -1712,12 +1690,12 @@ fn make_document(
     displayed_attributes: &BTreeSet<FieldId>,
     field_ids_map: &FieldsIdsMap,
     obkv: &obkv::KvReaderU16,
-) -> milli::Result<Document> {
+) -> Result<Document, MeilisearchHttpError> {
     let mut document = serde_json::Map::new();
 
     // recreate the original json
     for (key, value) in obkv.iter() {
-        let value = serde_json::from_slice(value).map_err(InternalError::SerdeJson)?;
+        let value = serde_json::from_slice(value)?;
         let key = field_ids_map.name(key).expect("Missing field name").to_string();
 
         document.insert(key, value);
@@ -1742,7 +1720,7 @@ fn format_fields(
     displayable_ids: &BTreeSet<FieldId>,
     locales: Option<&[Language]>,
     localized_attributes: &[LocalizedAttributesRule],
-) -> milli::Result<(Option<MatchesPosition>, Document)> {
+) -> Result<(Option<MatchesPosition>, Document), MeilisearchHttpError> {
     let mut matches_position = compute_matches.then(BTreeMap::new);
     let mut document = document.clone();
 
@@ -1920,7 +1898,7 @@ fn parse_filter_array(arr: &[Value]) -> Result<Option<Filter>, MeilisearchHttpEr
         }
     }
 
-    Filter::from_array(ands).map_err(|e| MeilisearchHttpError::from_milli(e, None))
+    Ok(Filter::from_array(ands)?)
 }
 
 #[cfg(test)]

crates/meilisearch/tests/batches/mod.rs

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

crates/meilisearch/tests/common/mod.rs

@@ -52,25 +52,6 @@ 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

@@ -1220,89 +1220,9 @@ async fn replace_document() {
 #[actix_rt::test]
 async fn add_no_documents() {
     let server = Server::new().await;
-    let index = server.index("kefir");
+    let index = server.index("test");
-    let (task, code) = index.add_documents(json!([]), None).await;
+    let (_response, 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]
@@ -1344,18 +1264,15 @@ 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"
         }
     ]);
-    let (value, _code) = index.add_documents(documents, None).await;
+    index.add_documents(documents, None).await;
-    index.wait_task(value.uid()).await;
+    index.wait_task(1).await;
-    let (response, code) = index.get_task(value.uid()).await;
+    let (response, code) = index.get_task(1).await;
     snapshot!(code, @"200 OK");
     snapshot!(json_string!(response, { ".duration" => "[duration]", ".enqueuedAt" => "[date]", ".startedAt" => "[date]", ".finishedAt" => "[date]" }),
         @r###"
@@ -1371,81 +1288,7 @@ 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 511 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 512 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"
@@ -1838,7 +1681,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "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\"`.",
+        "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\"`.",
         "code": "invalid_document_geo_field",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_document_geo_field"
@@ -1876,7 +1719,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Index `test`: Could not find latitude nor longitude in the document with the id: `\"11\"`. Was expecting `_geo.lat` and `_geo.lng` fields.",
+        "message": "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"
@@ -1914,7 +1757,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Index `test`: Could not find latitude nor longitude in the document with the id: `\"11\"`. Was expecting `_geo.lat` and `_geo.lng` fields.",
+        "message": "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"
@@ -1952,7 +1795,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Index `test`: Could not find longitude in the document with the id: `\"11\"`. Was expecting a `_geo.lng` field.",
+        "message": "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"
@@ -1990,7 +1833,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Index `test`: Could not find latitude in the document with the id: `\"11\"`. Was expecting a `_geo.lat` field.",
+        "message": "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"
@@ -2028,7 +1871,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Index `test`: Could not find longitude in the document with the id: `\"11\"`. Was expecting a `_geo.lng` field.",
+        "message": "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"
@@ -2066,7 +1909,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Index `test`: Could not find latitude in the document with the id: `\"11\"`. Was expecting a `_geo.lat` field.",
+        "message": "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"
@@ -2104,7 +1947,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "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`.",
+        "message": "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"
@@ -2142,7 +1985,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Index `test`: Could not find longitude in the document with the id: `\"11\"`. Was expecting a `_geo.lng` field.",
+        "message": "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"
@@ -2180,7 +2023,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Index `test`: Could not find latitude in the document with the id: `\"11\"`. Was expecting a `_geo.lat` field.",
+        "message": "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"
@@ -2218,7 +2061,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "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\"`.",
+        "message": "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"
@@ -2256,7 +2099,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Index `test`: The `_geo` field in the document with the id: `\"11\"` contains the following unexpected fields: `{\"doggo\":\"are the best\"}`.",
+        "message": "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"
@@ -2295,7 +2138,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Index `test`: Could not parse longitude in the document with the id: `\"12\"`. Was expecting a finite number but instead got `null`.",
+        "message": "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"
@@ -2332,7 +2175,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "message": "Index `test`: Could not parse latitude in the document with the id: `\"12\"`. Was expecting a finite number but instead got `null`.",
+        "message": "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"
@@ -2369,7 +2212,7 @@ async fn add_documents_invalid_geo_field() {
         "indexedDocuments": 0
       },
       "error": {
-        "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`.",
+        "message": "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"
@@ -2436,7 +2279,7 @@ async fn add_invalid_geo_and_then_settings() {
         ]
       },
       "error": {
-        "message": "Index `test`: Could not parse latitude in the document with the id: `\"11\"`. Was expecting a finite number but instead got `null`.",
+        "message": "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": "Index `EMPTY_INDEX`: Attribute `doggo` is not filterable. This index does not have configured filterable attributes.\n1:6 doggo = bernese",
+        "message": "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": "Index `SHARED_DOCUMENTS`: Attribute `catto` is not filterable. Available filterable attributes are: `id`, `title`.\n1:6 catto = jorts",
+        "message": "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 511 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 512 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 `test`: Index already has a primary key: `id`.",
+        "message": "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": format!("Index `{}`: Attribute `many` is not filterable. Available filterable attributes are: `title`.\n1:5 many = Glass", index.uid),
+        "message": "Attribute `many` is not filterable. Available filterable attributes are: `title`.\n1:5 many = Glass",
         "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": format!("Index `{}`: Attribute `many` is not filterable. Available filterable attributes are: `title`.\n1:5 many = Glass", index.uid),
+        "message": "Attribute `many` is not filterable. Available filterable attributes are: `title`.\n1:5 many = Glass",
         "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": format!("Index `{}`: Attribute `title` is not sortable. Available sortable attributes are: `id`.", index.uid),
+        "message": "Attribute `title` is not sortable. Available sortable attributes are: `id`.",
         "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": format!("Index `{}`: You must specify where `sort` is listed in the rankingRules setting to use the sort parameter at search time.", index.uid),
+        "message": "You must specify where `sort` is listed in the rankingRules setting to use the sort parameter at search time.",
         "code": "invalid_search_sort",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#invalid_search_sort"
@@ -1024,18 +1024,19 @@ 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| {
-                assert_eq!(response, expected_response);
+                snapshot!(code, @"400 Bad Request");
-                assert_eq!(code, 400);
+                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"
+                }
+                "###);
             },
         )
         .await;
@@ -1049,18 +1050,19 @@ 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| {
-                assert_eq!(response, expected_response);
+                snapshot!(code, @"400 Bad Request");
-                assert_eq!(code, 400);
+                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"
+                }
+                "###);
             },
         )
         .await;
@@ -1069,8 +1071,15 @@ async fn search_on_unknown_field_plus_joker() {
         .search(
             json!({"q": "Captain Marvel", "attributesToSearchOn": ["unknown", "*"]}),
             |response, code| {
-                assert_eq!(response, expected_response);
+                snapshot!(code, @"400 Bad Request");
-                assert_eq!(code, 400);
+                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"
+                }
+                "###);
             },
         )
         .await;
@@ -1083,44 +1092,47 @@ 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;
-    assert_eq!(response, expected_response);
+    snapshot!(code, @"400 Bad Request");
-    assert_eq!(code, 400);
+    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"
+    }
+    "###);
 
     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;
-    assert_eq!(response, expected_response);
+    snapshot!(code, @"400 Bad Request");
-    assert_eq!(code, 400);
+    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"
+    }
+    "###);
 
     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;
-    assert_eq!(response, expected_response);
+    snapshot!(code, @"400 Bad Request");
-    assert_eq!(code, 400);
+    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"
+    }
+    "###);
 
     let (response, code) =
         index.search_post(json!({"page": 0, "hitsPerPage": 2, "distinct": true})).await;

crates/meilisearch/tests/search/facet_search.rs

@@ -57,116 +57,6 @@ 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;
@@ -331,15 +221,8 @@ async fn add_documents_and_deactivate_facet_search() {
     let (response, code) =
         index.facet_search(json!({"facetName": "genres", "facetQuery": "a"})).await;
 
-    assert_eq!(code, 400, "{}", response);
+    assert_eq!(code, 200, "{}", response);
-    snapshot!(response, @r###"
+    assert_eq!(dbg!(response)["facetHits"].as_array().unwrap().len(), 0);
-    {
-      "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]
@@ -362,15 +245,8 @@ async fn deactivate_facet_search_and_add_documents() {
     let (response, code) =
         index.facet_search(json!({"facetName": "genres", "facetQuery": "a"})).await;
 
-    assert_eq!(code, 400, "{}", response);
+    assert_eq!(code, 200, "{}", response);
-    snapshot!(response, @r###"
+    assert_eq!(dbg!(response)["facetHits"].as_array().unwrap().len(), 0);
-    {
-      "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/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]`: Index `nested`: Attribute `title` is not filterable. This index does not have configured filterable attributes.\n1:6 title = toto",
+      "message": "Inside `.queries[1]`: 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]`: Index `nested`: Attribute `doggos` is not sortable. This index does not have configured sortable attributes.",
+      "message": "Inside `.queries[1]`: 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]`: Index `test`: Attribute `title` is not filterable. This index does not have configured filterable attributes.\n1:6 title = toto",
+      "message": "Inside `.queries[0]`: 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]`: Index `test`: Attribute `title` is not sortable. This index does not have configured sortable attributes.",
+      "message": "Inside `.queries[0]`: 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]`: Index `nested`: Attribute `doggos` is not filterable. This index does not have configured filterable attributes.\n1:7 doggos IN [intel, kefir]",
+      "message": "Inside `.queries[1]`: 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]`: Index `nested`: Attribute `doggos` is not sortable. This index does not have configured sortable attributes.",
+      "message": "Inside `.queries[1]`: 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/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 511 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 512 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 511 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 512 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;
 
-    let (task, _) = server.index("doggo").create(Some("bone")).await;
+    server.index("doggo").create(Some("bone")).await;
-    index.wait_task(task.uid()).await.succeeded();
+    index.wait_task(2).await;
 
-    let (task, _) = server.index("doggo").create(Some("bone")).await;
+    server.index("doggo").create(Some("bone")).await;
-    index.wait_task(task.uid()).await.failed();
+    index.wait_task(2).await;
 
     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": "Index `test`: Attribute `doggo` is not filterable. This index does not have configured filterable attributes.\n1:6 doggo = bernese",
+        "message": "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": "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.",
+        "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.",
         "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": "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`",
+        "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`",
         "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": "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`",
+        "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`",
         "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": "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",
+        "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",
         "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": "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`",
+        "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`",
         "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": "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`",
+        "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`",
         "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": "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`",
+        "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`",
         "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": "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]`",
+        "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]`",
         "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": "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`",
+        "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`",
         "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": "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`",
+        "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`",
         "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": "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`",
+        "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`",
         "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": "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).",
+        "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).",
         "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": "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).",
+        "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).",
         "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": "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",
+        "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",
         "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": "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",
+        "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",
         "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": "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",
+        "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",
         "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": "Index `doggo`: While embedding documents for embedder `rest`: runtime error: was expecting embeddings of dimension `2`, got embeddings of dimensions `3`",
+        "message": "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": "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\"}`",
+        "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\"}`",
         "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": "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\"}`",
+        "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\"}`",
         "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": "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)",
+        "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)",
         "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": "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",
+        "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",
         "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": "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",
+        "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",
         "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": "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",
+        "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",
         "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": "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?",
+        "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?",
         "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": "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",
+        "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",
         "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": "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",
+        "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",
         "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": "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\"}`",
+        "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\"}`",
         "code": "vector_embedding_error",
         "type": "invalid_request",
         "link": "https://docs.meilisearch.com/errors#vector_embedding_error"

crates/meilitool/Cargo.toml

@@ -10,15 +10,12 @@ 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 -> v1.12.x
+    /// - v1.9.x -> v1.10.x -> v1.11.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, detected_version)
+            export_a_dump(db_path, dump_dir, skip_enqueued_tasks)
         }
         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")?;
@@ -187,7 +187,6 @@ 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();
 
@@ -239,6 +238,9 @@ 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)? {
@@ -252,39 +254,18 @@ fn export_a_dump(
                 if status == Status::Enqueued {
                     let content_file = file_store.get_update(content_file_uuid)?;
 
-                    if (
+                    let reader =
-                        detected_version.0.as_str(),
+                        DocumentsBatchReader::from_reader(content_file).with_context(|| {
-                        detected_version.1.as_str(),
+                            format!("While reading content file {:?}", content_file_uuid)
-                        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)?;
-                        }
-                    }
 
+                    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)?)?;
+                    }
                     dump_content_file.flush()?;
                     count += 1;
                 }

crates/meilitool/src/upgrade/mod.rs

@@ -1,14 +1,13 @@
 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;
 
@@ -20,48 +19,11 @@ 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 = "1.12.5";
-        const FIRST_SUPPORTED_UPGRADE_TO_VERSION: &str = "1.10.0";
-        const LAST_SUPPORTED_UPGRADE_TO_VERSION: &str = "1.12.5";
-
         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 (
@@ -71,13 +33,8 @@ impl OfflineUpgrade {
         ) {
             ("1", "9", _) => 0,
             ("1", "10", _) => 1,
-            ("1", "11", _) => 2,
-            ("1", "12", "0" | "1" | "2") => 3,
-            ("1", "12", "3" | "4" | "5") => no_upgrade,
             _ => {
-                bail!("Unsupported current version {current_major}.{current_minor}.{current_patch}. Can only upgrade from versions in range [{}-{}]",
+                bail!("Unsupported current version {current_major}.{current_minor}.{current_patch}. Can only upgrade from v1.9 and v1.10")
-                      FIRST_SUPPORTED_UPGRADE_FROM_VERSION,
-                      LAST_SUPPORTED_UPGRADE_FROM_VERSION);
             }
         };
 
@@ -86,32 +43,20 @@ 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") => 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 versions in range [{}-{}]",
+                bail!("Unsupported target version {target_major}.{target_minor}.{target_patch}. Can only upgrade to v1.10 and v1.11")
-                      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, current_major, current_minor, current_patch)?;
+            (func)(&self.db_path)?;
             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

@@ -1,13 +1,18 @@
+use anyhow::bail;
 use std::path::Path;
 
-use anyhow::{bail, Context};
+use anyhow::Context;
-use meilisearch_types::heed::types::{SerdeJson, Str};
+use meilisearch_types::{
-use meilisearch_types::heed::{Database, Env, EnvOpenOptions, RoTxn, RwTxn, Unspecified};
+    heed::{
-use meilisearch_types::milli::index::{db_name, main_key};
+        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 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>;
 
@@ -151,12 +156,7 @@ fn date_round_trip(
     Ok(())
 }
 
-pub fn v1_9_to_v1_10(
+pub fn v1_9_to_v1_10(db_path: &Path) -> anyhow::Result<()> {
-    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

@@ -7,19 +7,14 @@
 use std::path::Path;
 
 use anyhow::Context;
-use meilisearch_types::heed::types::Str;
+use meilisearch_types::{
-use meilisearch_types::heed::{Database, EnvOpenOptions};
+    heed::{types::Str, Database, EnvOpenOptions},
-use meilisearch_types::milli::index::db_name;
+    milli::index::db_name,
+};
 
-use crate::uuid_codec::UuidCodec;
+use crate::{try_opening_database, try_opening_poly_database, uuid_codec::UuidCodec};
-use crate::{try_opening_database, try_opening_poly_database};
 
-pub fn v1_10_to_v1_11(
+pub fn v1_10_to_v1_11(db_path: &Path) -> anyhow::Result<()> {
-    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

@@ -1,298 +0,0 @@
-//! 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

@@ -91,8 +91,8 @@ 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"

crates/milli/src/documents/primary_key.rs

@@ -280,7 +280,7 @@ 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

@@ -114,7 +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 511 bytes.", .document_id.to_string()
+and can not be more than 512 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/index.rs

@@ -1734,7 +1734,6 @@ 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::{
@@ -1811,7 +1810,7 @@ pub(crate) mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                Progress::default(),
+                &|_progress| (),
             )?;
 
             if let Some(error) = operation_stats.into_iter().find_map(|stat| stat.error) {
@@ -1830,7 +1829,7 @@ pub(crate) mod tests {
                     &document_changes,
                     embedders,
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
             })
             .unwrap()?;
@@ -1902,7 +1901,7 @@ pub(crate) mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                Progress::default(),
+                &|_progress| (),
             )?;
 
             if let Some(error) = operation_stats.into_iter().find_map(|stat| stat.error) {
@@ -1921,7 +1920,7 @@ pub(crate) mod tests {
                     &document_changes,
                     embedders,
                     &|| false,
-                    &Progress::default(),
+                    &|_| (),
                 )
             })
             .unwrap()?;
@@ -1983,7 +1982,7 @@ pub(crate) mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                Progress::default(),
+                &|_progress| (),
             )
             .unwrap();
 
@@ -2002,7 +2001,7 @@ pub(crate) mod tests {
                     &document_changes,
                     embedders,
                     &|| should_abort.load(Relaxed),
-                    &Progress::default(),
+                    &|_| (),
                 )
             })
             .unwrap()

crates/milli/src/lib.rs

@@ -31,7 +31,6 @@ 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};

crates/milli/src/progress.rs

@@ -1,152 +0,0 @@
-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

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

@@ -38,16 +38,6 @@ 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 }
     }
@@ -57,8 +47,6 @@ 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

@@ -119,7 +119,6 @@ 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,
@@ -131,12 +130,9 @@ impl Prompt {
             self.max_bytes.unwrap_or_else(default_max_bytes).get(),
             doc_alloc,
         );
-        self.template.render_to(&mut rendered, &context).map_err(|liquid_error| {
+        self.template
-            RenderPromptError::missing_context_with_external_docid(
+            .render_to(&mut rendered, &context)
-                external_docid.to_owned(),
+            .map_err(RenderPromptError::missing_context)?;
-                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/new/tests/integration.rs

@@ -5,7 +5,6 @@ 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;
@@ -73,7 +72,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
             None,
             &mut new_fields_ids_map,
             &|| false,
-            Progress::default(),
+            &|_progress| (),
         )
         .unwrap();
 
@@ -92,7 +91,7 @@ pub fn setup_search_index_with_criteria(criteria: &[Criterion]) -> Index {
         &document_changes,
         embedders,
         &|| false,
-        &Progress::default(),
+        &|_| (),
     )
     .unwrap();
 

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

@@ -79,29 +79,22 @@ 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::{
+use crate::heed_codec::facet::{FacetGroupKey, FacetGroupKeyCodec, FacetGroupValueCodec};
-    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.
 ///
@@ -653,194 +646,3 @@ 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

@@ -1,498 +0,0 @@
-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/helpers/mod.rs

@@ -10,14 +10,10 @@ use fst::{IntoStreamer, Streamer};
 pub use grenad_helpers::*;
 pub use merge_functions::*;
 
-use crate::MAX_LMDB_KEY_LENGTH;
+use crate::MAX_WORD_LENGTH;
 
 pub fn valid_lmdb_key(key: impl AsRef<[u8]>) -> bool {
-    key.as_ref().len() <= MAX_LMDB_KEY_LENGTH - 3 && !key.as_ref().is_empty()
+    key.as_ref().len() <= MAX_WORD_LENGTH * 2 && !key.as_ref().is_empty()
-}
-
-pub fn valid_facet_value(facet_value: impl AsRef<[u8]>) -> bool {
-    facet_value.as_ref().len() <= MAX_LMDB_KEY_LENGTH - 3 && !facet_value.as_ref().is_empty()
 }
 
 /// Divides one slice into two at an index, returns `None` if mid is out of bounds.

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

@@ -766,7 +766,6 @@ mod tests {
     use crate::documents::mmap_from_objects;
     use crate::index::tests::TempIndex;
     use crate::index::IndexEmbeddingConfig;
-    use crate::progress::Progress;
     use crate::search::TermsMatchingStrategy;
     use crate::update::new::indexer;
     use crate::update::Setting;
@@ -1965,7 +1964,7 @@ mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                Progress::default(),
+                &|_progress| (),
             )
             .unwrap();
 
@@ -2149,7 +2148,7 @@ mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                Progress::default(),
+                &|_progress| (),
             )
             .unwrap();
 
@@ -2164,7 +2163,7 @@ mod tests {
             &document_changes,
             embedders,
             &|| false,
-            &Progress::default(),
+            &|_| (),
         )
         .unwrap();
         wtxn.commit().unwrap();
@@ -2211,7 +2210,7 @@ mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                Progress::default(),
+                &|_progress| (),
             )
             .unwrap();
 
@@ -2226,7 +2225,7 @@ mod tests {
             &document_changes,
             embedders,
             &|| false,
-            &Progress::default(),
+            &|_| (),
         )
         .unwrap();
         wtxn.commit().unwrap();
@@ -2264,7 +2263,7 @@ mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                Progress::default(),
+                &|_progress| (),
             )
             .unwrap();
 
@@ -2279,7 +2278,7 @@ mod tests {
             &document_changes,
             embedders,
             &|| false,
-            &Progress::default(),
+            &|_| (),
         )
         .unwrap();
         wtxn.commit().unwrap();
@@ -2316,7 +2315,7 @@ mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                Progress::default(),
+                &|_progress| (),
             )
             .unwrap();
 
@@ -2331,7 +2330,7 @@ mod tests {
             &document_changes,
             embedders,
             &|| false,
-            &Progress::default(),
+            &|_| (),
         )
         .unwrap();
         wtxn.commit().unwrap();
@@ -2370,7 +2369,7 @@ mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                Progress::default(),
+                &|_progress| (),
             )
             .unwrap();
 
@@ -2385,7 +2384,7 @@ mod tests {
             &document_changes,
             embedders,
             &|| false,
-            &Progress::default(),
+            &|_| (),
         )
         .unwrap();
         wtxn.commit().unwrap();
@@ -2429,7 +2428,7 @@ mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                Progress::default(),
+                &|_progress| (),
             )
             .unwrap();
 
@@ -2444,7 +2443,7 @@ mod tests {
             &document_changes,
             embedders,
             &|| false,
-            &Progress::default(),
+            &|_| (),
         )
         .unwrap();
         wtxn.commit().unwrap();
@@ -2481,7 +2480,7 @@ mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                Progress::default(),
+                &|_progress| (),
             )
             .unwrap();
 
@@ -2496,7 +2495,7 @@ mod tests {
             &document_changes,
             embedders,
             &|| false,
-            &Progress::default(),
+            &|_| (),
         )
         .unwrap();
         wtxn.commit().unwrap();
@@ -2533,7 +2532,7 @@ mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                Progress::default(),
+                &|_progress| (),
             )
             .unwrap();
 
@@ -2548,7 +2547,7 @@ mod tests {
             &document_changes,
             embedders,
             &|| false,
-            &Progress::default(),
+            &|_| (),
         )
         .unwrap();
         wtxn.commit().unwrap();
@@ -2727,7 +2726,7 @@ mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                Progress::default(),
+                &|_progress| (),
             )
             .unwrap();
 
@@ -2742,7 +2741,7 @@ mod tests {
             &document_changes,
             embedders,
             &|| false,
-            &Progress::default(),
+            &|_| (),
         )
         .unwrap();
         wtxn.commit().unwrap();
@@ -2786,7 +2785,7 @@ mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                Progress::default(),
+                &|_progress| (),
             )
             .unwrap();
 
@@ -2801,7 +2800,7 @@ mod tests {
             &document_changes,
             embedders,
             &|| false,
-            &Progress::default(),
+            &|_| (),
         )
         .unwrap();
         wtxn.commit().unwrap();
@@ -2842,7 +2841,7 @@ mod tests {
                 None,
                 &mut new_fields_ids_map,
                 &|| false,
-                Progress::default(),
+                &|_progress| (),
             )
             .unwrap();
 
@@ -2857,7 +2856,7 @@ mod tests {
             &document_changes,
             embedders,
             &|| false,
-            &Progress::default(),
+            &|_| (),
         )
         .unwrap();
         wtxn.commit().unwrap();
@@ -3334,44 +3333,6 @@ mod tests {
         rtxn.commit().unwrap();
     }
 
-    #[test]
-    fn incremental_update_without_changing_facet_distribution() {
-        let index = TempIndex::new();
-        index
-            .add_documents(documents!([
-                {"id": 0, "some_field": "aaa", "other_field": "aaa" },
-                {"id": 1, "some_field": "bbb", "other_field": "bbb" },
-            ]))
-            .unwrap();
-        {
-            let rtxn = index.read_txn().unwrap();
-            // count field distribution
-            let results = index.field_distribution(&rtxn).unwrap();
-            assert_eq!(Some(&2), results.get("id"));
-            assert_eq!(Some(&2), results.get("some_field"));
-            assert_eq!(Some(&2), results.get("other_field"));
-        }
-
-        let mut index = index;
-        index.index_documents_config.update_method = IndexDocumentsMethod::UpdateDocuments;
-
-        index
-            .add_documents(documents!([
-                {"id": 0, "other_field": "bbb" },
-                {"id": 1, "some_field": "ccc" },
-            ]))
-            .unwrap();
-
-        {
-            let rtxn = index.read_txn().unwrap();
-            // count field distribution
-            let results = index.field_distribution(&rtxn).unwrap();
-            assert_eq!(Some(&2), results.get("id"));
-            assert_eq!(Some(&2), results.get("some_field"));
-            assert_eq!(Some(&2), results.get("other_field"));
-        }
-    }
-
     #[test]
     fn delete_words_exact_attributes() {
         let index = TempIndex::new();

crates/milli/src/update/index_documents/snapshots/mod.rs/geo_filtered_placeholder_search_should_not_return_deleted_documents/facet_id_f64_docids.snap

@@ -1,5 +1,5 @@
 ---
-source: crates/milli/src/update/index_documents/mod.rs
+source: milli/src/update/index_documents/mod.rs
 ---
 3   0  48.9021 1  [19, ]
 3   0  49.9314 1  [17, ]
@@ -15,11 +15,6 @@ source: crates/milli/src/update/index_documents/mod.rs
 3   0  50.7453 1  [7, ]
 3   0  50.8466 1  [10, ]
 3   0  51.0537 1  [9, ]
-3   1  48.9021 2  [17, 19, ]
-3   1  50.1793 3  [13, 14, 15, ]
-3   1  50.4502 4  [0, 3, 8, 12, ]
-3   1  50.6312 2  [1, 2, ]
-3   1  50.7453 3  [7, 9, 10, ]
 4   0  2.271  1  [17, ]
 4   0  2.3708 1  [19, ]
 4   0  2.7637 1  [14, ]
@@ -33,3 +28,4 @@ source: crates/milli/src/update/index_documents/mod.rs
 4   0  3.6957 1  [9, ]
 4   0  3.9623 1  [12, ]
 4   0  4.337  1  [10, ]
+

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

@@ -1,8 +1,7 @@
 use std::collections::{BTreeMap, BTreeSet};
 
-use bumparaw_collections::RawMap;
 use heed::RoTxn;
-use rustc_hash::FxBuildHasher;
+use raw_collections::RawMap;
 use serde_json::value::RawValue;
 
 use super::vector_document::VectorDocument;
@@ -386,12 +385,12 @@ pub type Entry<'doc> = (&'doc str, &'doc RawValue);
 
 #[derive(Debug)]
 pub struct Versions<'doc> {
-    data: RawMap<'doc, FxBuildHasher>,
+    data: RawMap<'doc>,
 }
 
 impl<'doc> Versions<'doc> {
     pub fn multiple(
-        mut versions: impl Iterator<Item = Result<RawMap<'doc, FxBuildHasher>>>,
+        mut versions: impl Iterator<Item = Result<RawMap<'doc>>>,
     ) -> Result<Option<Self>> {
         let Some(data) = versions.next() else { return Ok(None) };
         let mut data = data?;
@@ -404,7 +403,7 @@ impl<'doc> Versions<'doc> {
         Ok(Some(Self::single(data)))
     }
 
-    pub fn single(version: RawMap<'doc, FxBuildHasher>) -> Self {
+    pub fn single(version: RawMap<'doc>) -> Self {
         Self { data: version }
     }
 

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

@@ -1,10 +1,7 @@
 use bumpalo::Bump;
 use heed::RoTxn;
 
-use super::document::{
+use super::document::{DocumentFromDb, DocumentFromVersions, MergedDocument, Versions};
-    Document as _, DocumentFromDb, DocumentFromVersions, MergedDocument, Versions,
-};
-use super::extract::perm_json_p;
 use super::vector_document::{
     MergedVectorDocument, VectorDocumentFromDb, VectorDocumentFromVersions,
 };
@@ -167,80 +164,6 @@ impl<'doc> Update<'doc> {
         }
     }
 
-    /// Returns whether the updated version of the document is different from the current version for the passed subset of fields.
-    ///
-    /// `true` if at least one top-level-field that is a exactly a member of field or a parent of a member of field changed.
-    /// Otherwise `false`.
-    pub fn has_changed_for_fields<'t, Mapper: FieldIdMapper>(
-        &self,
-        fields: Option<&[&str]>,
-        rtxn: &'t RoTxn,
-        index: &'t Index,
-        mapper: &'t Mapper,
-    ) -> Result<bool> {
-        let mut changed = false;
-        let mut cached_current = None;
-        let mut updated_selected_field_count = 0;
-
-        for entry in self.updated().iter_top_level_fields() {
-            let (key, updated_value) = entry?;
-
-            if perm_json_p::select_field(key, fields, &[]) == perm_json_p::Selection::Skip {
-                continue;
-            }
-
-            updated_selected_field_count += 1;
-            let current = match cached_current {
-                Some(current) => current,
-                None => self.current(rtxn, index, mapper)?,
-            };
-            let current_value = current.top_level_field(key)?;
-            let Some(current_value) = current_value else {
-                changed = true;
-                break;
-            };
-
-            if current_value.get() != updated_value.get() {
-                changed = true;
-                break;
-            }
-            cached_current = Some(current);
-        }
-
-        if !self.has_deletion {
-            // no field deletion, so fields that don't appear in `updated` cannot have changed
-            return Ok(changed);
-        }
-
-        if changed {
-            return Ok(true);
-        }
-
-        // we saw all updated fields, and set `changed` if any field wasn't in `current`.
-        // so if there are as many fields in `current` as in `updated`, then nothing changed.
-        // If there is any more fields in `current`, then they are missing in `updated`.
-        let has_deleted_fields = {
-            let current = match cached_current {
-                Some(current) => current,
-                None => self.current(rtxn, index, mapper)?,
-            };
-
-            let mut current_selected_field_count = 0;
-            for entry in current.iter_top_level_fields() {
-                let (key, _) = entry?;
-
-                if perm_json_p::select_field(key, fields, &[]) == perm_json_p::Selection::Skip {
-                    continue;
-                }
-                current_selected_field_count += 1;
-            }
-
-            current_selected_field_count != updated_selected_field_count
-        };
-
-        Ok(has_deleted_fields)
-    }
-
     pub fn updated_vectors(
         &self,
         doc_alloc: &'doc Bump,

crates/milli/src/update/new/extract/cache.rs

@@ -60,21 +60,19 @@
 //! For now we can use a grenad sorter for spilling even thought I think
 //! it's not the most efficient way (too many files open, sorting entries).
 
+use std::borrow::Borrow;
 use std::cmp::Ordering;
 use std::collections::binary_heap::PeekMut;
-use std::collections::BinaryHeap;
+use std::collections::{BTreeMap, BinaryHeap};
 use std::fs::File;
 use std::hash::BuildHasher;
 use std::io::BufReader;
 use std::{io, iter, mem};
 
 use bumpalo::Bump;
-use bumparaw_collections::bbbul::{BitPacker, BitPacker4x};
-use bumparaw_collections::map::FrozenMap;
-use bumparaw_collections::{Bbbul, FrozenBbbul};
 use grenad::ReaderCursor;
-use hashbrown::hash_map::RawEntryMut;
+use raw_collections::bbbul::{BitPacker, BitPacker4x};
-use hashbrown::HashMap;
+use raw_collections::{Bbbul, FrozenBbbul};
 use roaring::RoaringBitmap;
 use rustc_hash::FxBuildHasher;
 
@@ -105,9 +103,7 @@ impl<'extractor> BalancedCaches<'extractor> {
             hasher: FxBuildHasher,
             max_memory,
             caches: InnerCaches::Normal(NormalCaches {
-                caches: iter::repeat_with(|| HashMap::with_hasher_in(FxBuildHasher, alloc))
+                caches: iter::repeat_with(BTreeMap::new).take(buckets).collect(),
-                    .take(buckets)
-                    .collect(),
             }),
             alloc,
         }
@@ -166,8 +162,8 @@ impl<'extractor> BalancedCaches<'extractor> {
                 rayon::current_thread_index().unwrap_or(0)
             );
 
-            let allocated: usize = normal_caches.caches.iter().map(|m| m.allocation_size()).sum();
+            // let allocated: usize = normal_caches.caches.iter().map(|m| m.allocation_size()).sum();
-            tracing::trace!("The last allocated HashMap took {allocated} bytes");
+            // tracing::trace!("The last allocated BTreeMap took {allocated} bytes");
 
             let dummy = NormalCaches { caches: Vec::new() };
             let NormalCaches { caches: cache_maps } = mem::replace(normal_caches, dummy);
@@ -177,43 +173,34 @@ impl<'extractor> BalancedCaches<'extractor> {
         Ok(())
     }
 
-    pub fn freeze(&mut self, source_id: usize) -> Result<Vec<FrozenCache<'_, 'extractor>>> {
+    pub fn freeze(&mut self) -> Result<Vec<FrozenCache<'_, 'extractor>>> {
         match &mut self.caches {
             InnerCaches::Normal(NormalCaches { caches }) => caches
                 .iter_mut()
                 .enumerate()
-                .map(|(bucket_id, map)| {
+                .map(|(bucket, map)| {
                     // safety: we are transmuting the Bbbul into a FrozenBbbul
                     //         that are the same size.
                     let map = unsafe {
                         std::mem::transmute::<
-                            &mut HashMap<
+                            &mut BTreeMap<
                                 &[u8],
                                 DelAddBbbul<BitPacker4x>, // from this
-                                FxBuildHasher,
-                                &Bump,
                             >,
-                            &mut HashMap<
+                            &mut BTreeMap<
                                 &[u8],
                                 FrozenDelAddBbbul<BitPacker4x>, // to that
-                                FxBuildHasher,
-                                &Bump,
                             >,
                         >(map)
                     };
-                    Ok(FrozenCache {
+                    Ok(FrozenCache { bucket, cache: FrozenBTreeMap::new(map), spilled: Vec::new() })
-                        source_id,
-                        bucket_id,
-                        cache: FrozenMap::new(map),
-                        spilled: Vec::new(),
-                    })
                 })
                 .collect(),
             InnerCaches::Spilling(SpillingCaches { caches, spilled_entries, .. }) => caches
                 .iter_mut()
                 .zip(mem::take(spilled_entries))
                 .enumerate()
-                .map(|(bucket_id, (map, sorter))| {
+                .map(|(bucket, (map, sorter))| {
                     let spilled = sorter
                         .into_reader_cursors()?
                         .into_iter()
@@ -225,21 +212,17 @@ impl<'extractor> BalancedCaches<'extractor> {
                     //         that are the same size.
                     let map = unsafe {
                         std::mem::transmute::<
-                            &mut HashMap<
+                            &mut BTreeMap<
                                 &[u8],
                                 DelAddBbbul<BitPacker4x>, // from this
-                                FxBuildHasher,
-                                &Bump,
                             >,
-                            &mut HashMap<
+                            &mut BTreeMap<
                                 &[u8],
                                 FrozenDelAddBbbul<BitPacker4x>, // to that
-                                FxBuildHasher,
-                                &Bump,
                             >,
                         >(map)
                     };
-                    Ok(FrozenCache { source_id, bucket_id, cache: FrozenMap::new(map), spilled })
+                    Ok(FrozenCache { bucket, cache: FrozenBTreeMap::new(map), spilled })
                 })
                 .collect(),
         }
@@ -250,14 +233,7 @@ impl<'extractor> BalancedCaches<'extractor> {
 unsafe impl MostlySend for BalancedCaches<'_> {}
 
 struct NormalCaches<'extractor> {
-    caches: Vec<
+    caches: Vec<BTreeMap<&'extractor [u8], DelAddBbbul<'extractor, BitPacker4x>>>,
-        HashMap<
-            &'extractor [u8],
-            DelAddBbbul<'extractor, BitPacker4x>,
-            FxBuildHasher,
-            &'extractor Bump,
-        >,
-    >,
 }
 
 impl<'extractor> NormalCaches<'extractor> {
@@ -271,17 +247,13 @@ impl<'extractor> NormalCaches<'extractor> {
     ) {
         let hash = hasher.hash_one(key);
         let bucket = compute_bucket_from_hash(buckets, hash);
-
+        let cache = &mut self.caches[bucket];
-        match self.caches[bucket].raw_entry_mut().from_hash(hash, |&k| k == key) {
+        match cache.get_mut(key) {
-            RawEntryMut::Occupied(mut entry) => {
+            Some(deladd) => {
-                entry.get_mut().del.get_or_insert_with(|| Bbbul::new_in(alloc)).insert(n);
+                deladd.del.get_or_insert_with(|| Bbbul::new_in(alloc)).insert(n);
             }
-            RawEntryMut::Vacant(entry) => {
+            None => {
-                entry.insert_hashed_nocheck(
+                cache.insert(alloc.alloc_slice_copy(key), DelAddBbbul::new_del_u32_in(n, alloc));
-                    hash,
-                    alloc.alloc_slice_copy(key),
-                    DelAddBbbul::new_del_u32_in(n, alloc),
-                );
             }
         }
     }
@@ -296,30 +268,20 @@ impl<'extractor> NormalCaches<'extractor> {
     ) {
         let hash = hasher.hash_one(key);
         let bucket = compute_bucket_from_hash(buckets, hash);
-        match self.caches[bucket].raw_entry_mut().from_hash(hash, |&k| k == key) {
+        let cache = &mut self.caches[bucket];
-            RawEntryMut::Occupied(mut entry) => {
+        match cache.get_mut(key) {
-                entry.get_mut().add.get_or_insert_with(|| Bbbul::new_in(alloc)).insert(n);
+            Some(deladd) => {
+                deladd.add.get_or_insert_with(|| Bbbul::new_in(alloc)).insert(n);
             }
-            RawEntryMut::Vacant(entry) => {
+            None => {
-                entry.insert_hashed_nocheck(
+                cache.insert(alloc.alloc_slice_copy(key), DelAddBbbul::new_add_u32_in(n, alloc));
-                    hash,
-                    alloc.alloc_slice_copy(key),
-                    DelAddBbbul::new_add_u32_in(n, alloc),
-                );
             }
         }
     }
 }
 
 struct SpillingCaches<'extractor> {
-    caches: Vec<
+    caches: Vec<BTreeMap<&'extractor [u8], DelAddBbbul<'extractor, BitPacker4x>>>,
-        HashMap<
-            &'extractor [u8],
-            DelAddBbbul<'extractor, BitPacker4x>,
-            FxBuildHasher,
-            &'extractor Bump,
-        >,
-    >,
     spilled_entries: Vec<grenad::Sorter<MergeDeladdCboRoaringBitmaps>>,
     deladd_buffer: Vec<u8>,
     cbo_buffer: Vec<u8>,
@@ -327,14 +289,7 @@ struct SpillingCaches<'extractor> {
 
 impl<'extractor> SpillingCaches<'extractor> {
     fn from_cache_maps(
-        caches: Vec<
+        caches: Vec<BTreeMap<&'extractor [u8], DelAddBbbul<'extractor, BitPacker4x>>>,
-            HashMap<
-                &'extractor [u8],
-                DelAddBbbul<'extractor, BitPacker4x>,
-                FxBuildHasher,
-                &'extractor Bump,
-            >,
-        >,
     ) -> SpillingCaches<'extractor> {
         SpillingCaches {
             spilled_entries: iter::repeat_with(|| {
@@ -361,12 +316,12 @@ impl<'extractor> SpillingCaches<'extractor> {
     ) -> Result<()> {
         let hash = hasher.hash_one(key);
         let bucket = compute_bucket_from_hash(buckets, hash);
-        match self.caches[bucket].raw_entry_mut().from_hash(hash, |&k| k == key) {
+        match self.caches[bucket].get_mut(key) {
-            RawEntryMut::Occupied(mut entry) => {
+            Some(deladd) => {
-                entry.get_mut().del.get_or_insert_with(|| Bbbul::new_in(alloc)).insert(n);
+                deladd.del.get_or_insert_with(|| Bbbul::new_in(alloc)).insert(n);
                 Ok(())
             }
-            RawEntryMut::Vacant(_entry) => spill_entry_to_sorter(
+            None => spill_entry_to_sorter(
                 &mut self.spilled_entries[bucket],
                 &mut self.deladd_buffer,
                 &mut self.cbo_buffer,
@@ -386,12 +341,12 @@ impl<'extractor> SpillingCaches<'extractor> {
     ) -> Result<()> {
         let hash = hasher.hash_one(key);
         let bucket = compute_bucket_from_hash(buckets, hash);
-        match self.caches[bucket].raw_entry_mut().from_hash(hash, |&k| k == key) {
+        match self.caches[bucket].get_mut(key) {
-            RawEntryMut::Occupied(mut entry) => {
+            Some(deladd) => {
-                entry.get_mut().add.get_or_insert_with(|| Bbbul::new_in(alloc)).insert(n);
+                deladd.add.get_or_insert_with(|| Bbbul::new_in(alloc)).insert(n);
                 Ok(())
             }
-            RawEntryMut::Vacant(_entry) => spill_entry_to_sorter(
+            None => spill_entry_to_sorter(
                 &mut self.spilled_entries[bucket],
                 &mut self.deladd_buffer,
                 &mut self.cbo_buffer,
@@ -445,15 +400,8 @@ fn spill_entry_to_sorter(
 }
 
 pub struct FrozenCache<'a, 'extractor> {
-    bucket_id: usize,
+    bucket: usize,
-    source_id: usize,
+    cache: FrozenBTreeMap<'a, &'extractor [u8], FrozenDelAddBbbul<'extractor, BitPacker4x>>,
-    cache: FrozenMap<
-        'a,
-        'extractor,
-        &'extractor [u8],
-        FrozenDelAddBbbul<'extractor, BitPacker4x>,
-        FxBuildHasher,
-    >,
     spilled: Vec<grenad::Reader<BufReader<File>>>,
 }
 
@@ -463,15 +411,45 @@ pub fn transpose_and_freeze_caches<'a, 'extractor>(
     let width = caches.first().map(BalancedCaches::buckets).unwrap_or(0);
     let mut bucket_caches: Vec<_> = iter::repeat_with(Vec::new).take(width).collect();
 
-    for (thread_index, thread_cache) in caches.iter_mut().enumerate() {
+    for thread_cache in caches {
-        for frozen in thread_cache.freeze(thread_index)? {
+        for frozen in thread_cache.freeze()? {
-            bucket_caches[frozen.bucket_id].push(frozen);
+            bucket_caches[frozen.bucket].push(frozen);
         }
     }
 
     Ok(bucket_caches)
 }
 
+pub struct FrozenBTreeMap<'a, K, V>(&'a mut BTreeMap<K, V>);
+
+unsafe impl<'a, K, V> Send for FrozenBTreeMap<'a, K, V>
+where
+    K: Send,
+    V: Send,
+{
+}
+
+impl<'a, K, V> FrozenBTreeMap<'a, K, V> {
+    #[inline]
+    pub fn new(map: &'a mut BTreeMap<K, V>) -> Self {
+        Self(map)
+    }
+
+    #[inline]
+    pub fn iter_mut(&mut self) -> std::collections::btree_map::IterMut<'_, K, V> {
+        self.0.iter_mut()
+    }
+
+    #[inline]
+    pub fn get_mut<Q>(&mut self, key: &Q) -> Option<&mut V>
+    where
+        K: Borrow<Q> + Ord,
+        Q: Ord + ?Sized,
+    {
+        self.0.get_mut(key)
+    }
+}
+
 /// Merges the caches that must be all associated to the same bucket
 /// but make sure to sort the different buckets before performing the merges.
 ///
@@ -483,16 +461,21 @@ where
     F: for<'a> FnMut(&'a [u8], DelAddRoaringBitmap) -> Result<()>,
 {
     let mut maps = Vec::new();
-    let mut heap = BinaryHeap::new();
+    let mut readers = Vec::new();
     let mut current_bucket = None;
-    for FrozenCache { source_id, bucket_id, cache, spilled } in frozen {
+    for FrozenCache { bucket, cache, ref mut spilled } in frozen {
-        assert_eq!(*current_bucket.get_or_insert(bucket_id), bucket_id);
+        assert_eq!(*current_bucket.get_or_insert(bucket), bucket);
-        maps.push((source_id, cache));
+        maps.push(cache);
-        for reader in spilled {
+        readers.append(spilled);
-            let mut cursor = reader.into_cursor()?;
+    }
-            if cursor.move_on_next()?.is_some() {
+
-                heap.push(Entry { cursor, source_id });
+    // First manage the spilled entries by looking into the HashMaps,
-            }
+    // merge them and mark them as dummy.
+    let mut heap = BinaryHeap::new();
+    for (source_index, source) in readers.into_iter().enumerate() {
+        let mut cursor = source.into_cursor()?;
+        if cursor.move_on_next()?.is_some() {
+            heap.push(CursorEntry { cursor, source_index });
         }
     }
 
@@ -509,29 +492,25 @@ where
 
         let mut output = DelAddRoaringBitmap::from_bytes(first_value)?;
         while let Some(mut entry) = heap.peek_mut() {
-            if let Some((key, value)) = entry.cursor.current() {
+            if let Some((key, _value)) = entry.cursor.current() {
-                if first_key != key {
+                if first_key == key {
+                    let new = DelAddRoaringBitmap::from_bytes(first_value)?;
+                    output = output.merge(new);
+                    // When we are done we the current value of this entry move make
+                    // it move forward and let the heap reorganize itself (on drop)
+                    if entry.cursor.move_on_next()?.is_none() {
+                        PeekMut::pop(entry);
+                    }
+                } else {
                     break;
                 }
-
-                let new = DelAddRoaringBitmap::from_bytes(value)?;
-                output = output.merge(new);
-                // When we are done we the current value of this entry move make
-                // it move forward and let the heap reorganize itself (on drop)
-                if entry.cursor.move_on_next()?.is_none() {
-                    PeekMut::pop(entry);
-                }
             }
         }
 
         // Once we merged all of the spilled bitmaps we must also
         // fetch the entries from the non-spilled entries (the HashMaps).
-        for (source_id, map) in maps.iter_mut() {
+        for (map_index, map) in maps.iter_mut().enumerate() {
-            debug_assert!(
+            if first_entry.source_index != map_index {
-                !(map.get(first_key).is_some() && first_entry.source_id == *source_id),
-                "A thread should not have spiled a key that has been inserted in the cache"
-            );
-            if first_entry.source_id != *source_id {
                 if let Some(new) = map.get_mut(first_key) {
                     output.union_and_clear_bbbul(new);
                 }
@@ -543,22 +522,21 @@ where
 
         // Don't forget to put the first entry back into the heap.
         if first_entry.cursor.move_on_next()?.is_some() {
-            heap.push(first_entry);
+            heap.push(first_entry)
         }
     }
 
     // Then manage the content on the HashMap entries that weren't taken (mem::take).
-    while let Some((_, mut map)) = maps.pop() {
+    while let Some(mut map) = maps.pop() {
-        // Make sure we don't try to work with entries already managed by the spilled
+        for (key, bbbul) in map.iter_mut() {
-        let mut ordered_entries: Vec<_> =
+            if bbbul.is_empty() {
-            map.iter_mut().filter(|(_, bbbul)| !bbbul.is_empty()).collect();
+                continue;
-        ordered_entries.sort_unstable_by_key(|(key, _)| *key);
+            }
 
-        for (key, bbbul) in ordered_entries {
             let mut output = DelAddRoaringBitmap::empty();
             output.union_and_clear_bbbul(bbbul);
 
-            for (_, rhs) in maps.iter_mut() {
+            for rhs in maps.iter_mut() {
                 if let Some(new) = rhs.get_mut(key) {
                     output.union_and_clear_bbbul(new);
                 }
@@ -572,29 +550,29 @@ where
     Ok(())
 }
 
-struct Entry<R> {
+struct CursorEntry<R> {
     cursor: ReaderCursor<R>,
-    source_id: usize,
+    source_index: usize,
 }
 
-impl<R> Ord for Entry<R> {
+impl<R> Ord for CursorEntry<R> {
-    fn cmp(&self, other: &Entry<R>) -> Ordering {
+    fn cmp(&self, other: &CursorEntry<R>) -> Ordering {
         let skey = self.cursor.current().map(|(k, _)| k);
         let okey = other.cursor.current().map(|(k, _)| k);
-        skey.cmp(&okey).then(self.source_id.cmp(&other.source_id)).reverse()
+        skey.cmp(&okey).then(self.source_index.cmp(&other.source_index)).reverse()
     }
 }
 
-impl<R> Eq for Entry<R> {}
+impl<R> Eq for CursorEntry<R> {}
 
-impl<R> PartialEq for Entry<R> {
+impl<R> PartialEq for CursorEntry<R> {
-    fn eq(&self, other: &Entry<R>) -> bool {
+    fn eq(&self, other: &CursorEntry<R>) -> bool {
         self.cmp(other) == Ordering::Equal
     }
 }
 
-impl<R> PartialOrd for Entry<R> {
+impl<R> PartialOrd for CursorEntry<R> {
-    fn partial_cmp(&self, other: &Entry<R>) -> Option<Ordering> {
+    fn partial_cmp(&self, other: &CursorEntry<R>) -> Option<Ordering> {
         Some(self.cmp(other))
     }
 }

crates/milli/src/update/new/extract/documents.rs

@@ -89,8 +89,7 @@ impl<'a, 'b, 'extractor> Extractor<'extractor> for DocumentsExtractor<'a, 'b> {
                             .or_default();
                         *entry -= 1;
                     }
-                    let content =
+                    let content = update.updated();
-                        update.merged(&context.rtxn, context.index, &context.db_fields_ids_map)?;
                     let geo_iter =
                         content.geo_field().transpose().map(|res| res.map(|rv| ("_geo", rv)));
                     for res in content.iter_top_level_fields().chain(geo_iter) {

crates/milli/src/update/new/extract/faceted/extract_facets.rs

@@ -16,10 +16,10 @@ use crate::update::del_add::DelAdd;
 use crate::update::new::channel::FieldIdDocidFacetSender;
 use crate::update::new::extract::perm_json_p;
 use crate::update::new::indexer::document_changes::{
-    extract, DocumentChangeContext, DocumentChanges, Extractor, IndexingContext,
+    extract, DocumentChangeContext, DocumentChanges, Extractor, IndexingContext, Progress,
 };
 use crate::update::new::ref_cell_ext::RefCellExt as _;
-use crate::update::new::steps::IndexingStep;
+use crate::update::new::steps::Step;
 use crate::update::new::thread_local::{FullySend, ThreadLocal};
 use crate::update::new::DocumentChange;
 use crate::update::GrenadParameters;
@@ -97,15 +97,6 @@ impl FacetedDocidsExtractor {
                 },
             ),
             DocumentChange::Update(inner) => {
-                if !inner.has_changed_for_fields(
-                    Some(attributes_to_extract),
-                    rtxn,
-                    index,
-                    context.db_fields_ids_map,
-                )? {
-                    return Ok(());
-                }
-
                 extract_document_facets(
                     attributes_to_extract,
                     inner.current(rtxn, index, context.db_fields_ids_map)?,
@@ -373,16 +364,26 @@ fn truncate_str(s: &str) -> &str {
 
 impl FacetedDocidsExtractor {
     #[tracing::instrument(level = "trace", skip_all, target = "indexing::extract::faceted")]
-    pub fn run_extraction<'pl, 'fid, 'indexer, 'index, 'extractor, DC: DocumentChanges<'pl>, MSP>(
+    pub fn run_extraction<
+        'pl,
+        'fid,
+        'indexer,
+        'index,
+        'extractor,
+        DC: DocumentChanges<'pl>,
+        MSP,
+        SP,
+    >(
         grenad_parameters: GrenadParameters,
         document_changes: &DC,
-        indexing_context: IndexingContext<'fid, 'indexer, 'index, MSP>,
+        indexing_context: IndexingContext<'fid, 'indexer, 'index, MSP, SP>,
         extractor_allocs: &'extractor mut ThreadLocal<FullySend<Bump>>,
         sender: &FieldIdDocidFacetSender,
-        step: IndexingStep,
+        step: Step,
     ) -> Result<Vec<BalancedCaches<'extractor>>>
     where
         MSP: Fn() -> bool + Sync,
+        SP: Fn(Progress) + Sync,
     {
         let index = indexing_context.index;
         let rtxn = index.read_txn()?;

crates/milli/src/update/new/extract/mod.rs

@@ -15,22 +15,23 @@ pub use geo::*;
 pub use searchable::*;
 pub use vectors::EmbeddingExtractor;
 
-use super::indexer::document_changes::{DocumentChanges, IndexingContext};
+use super::indexer::document_changes::{DocumentChanges, IndexingContext, Progress};
-use super::steps::IndexingStep;
+use super::steps::Step;
 use super::thread_local::{FullySend, ThreadLocal};
 use crate::update::GrenadParameters;
 use crate::Result;
 
 pub trait DocidsExtractor {
-    fn run_extraction<'pl, 'fid, 'indexer, 'index, 'extractor, DC: DocumentChanges<'pl>, MSP>(
+    fn run_extraction<'pl, 'fid, 'indexer, 'index, 'extractor, DC: DocumentChanges<'pl>, MSP, SP>(
         grenad_parameters: GrenadParameters,
         document_changes: &DC,
-        indexing_context: IndexingContext<'fid, 'indexer, 'index, MSP>,
+        indexing_context: IndexingContext<'fid, 'indexer, 'index, MSP, SP>,
         extractor_allocs: &'extractor mut ThreadLocal<FullySend<Bump>>,
-        step: IndexingStep,
+        step: Step,
     ) -> Result<Vec<BalancedCaches<'extractor>>>
     where
-        MSP: Fn() -> bool + Sync;
+        MSP: Fn() -> bool + Sync,
+        SP: Fn(Progress) + Sync;
 }
 
 /// TODO move in permissive json pointer

crates/milli/src/update/new/extract/searchable/extract_word_docids.rs

@@ -11,10 +11,10 @@ use super::tokenize_document::{tokenizer_builder, DocumentTokenizer};
 use crate::update::new::extract::cache::BalancedCaches;
 use crate::update::new::extract::perm_json_p::contained_in;
 use crate::update::new::indexer::document_changes::{
-    extract, DocumentChangeContext, DocumentChanges, Extractor, IndexingContext,
+    extract, DocumentChangeContext, DocumentChanges, Extractor, IndexingContext, Progress,
 };
 use crate::update::new::ref_cell_ext::RefCellExt as _;
-use crate::update::new::steps::IndexingStep;
+use crate::update::new::steps::Step;
 use crate::update::new::thread_local::{FullySend, MostlySend, ThreadLocal};
 use crate::update::new::DocumentChange;
 use crate::update::GrenadParameters;
@@ -28,7 +28,7 @@ pub struct WordDocidsBalancedCaches<'extractor> {
     exact_word_docids: BalancedCaches<'extractor>,
     word_position_docids: BalancedCaches<'extractor>,
     fid_word_count_docids: BalancedCaches<'extractor>,
-    fid_word_count: HashMap<FieldId, (Option<usize>, Option<usize>)>,
+    fid_word_count: HashMap<FieldId, (usize, usize)>,
     current_docid: Option<DocumentId>,
 }
 
@@ -85,8 +85,8 @@ impl<'extractor> WordDocidsBalancedCaches<'extractor> {
 
         self.fid_word_count
             .entry(field_id)
-            .and_modify(|(_current_count, new_count)| *new_count.get_or_insert(0) += 1)
+            .and_modify(|(_current_count, new_count)| *new_count += 1)
-            .or_insert((None, Some(1)));
+            .or_insert((0, 1));
         self.current_docid = Some(docid);
 
         Ok(())
@@ -130,8 +130,8 @@ impl<'extractor> WordDocidsBalancedCaches<'extractor> {
 
         self.fid_word_count
             .entry(field_id)
-            .and_modify(|(current_count, _new_count)| *current_count.get_or_insert(0) += 1)
+            .and_modify(|(current_count, _new_count)| *current_count += 1)
-            .or_insert((Some(1), None));
+            .or_insert((1, 0));
 
         self.current_docid = Some(docid);
 
@@ -141,18 +141,14 @@ impl<'extractor> WordDocidsBalancedCaches<'extractor> {
     fn flush_fid_word_count(&mut self, buffer: &mut BumpVec<u8>) -> Result<()> {
         for (fid, (current_count, new_count)) in self.fid_word_count.drain() {
             if current_count != new_count {
-                if let Some(current_count) =
+                if current_count <= MAX_COUNTED_WORDS {
-                    current_count.filter(|current_count| *current_count <= MAX_COUNTED_WORDS)
-                {
                     buffer.clear();
                     buffer.extend_from_slice(&fid.to_be_bytes());
                     buffer.push(current_count as u8);
                     self.fid_word_count_docids
                         .insert_del_u32(buffer, self.current_docid.unwrap())?;
                 }
-                if let Some(new_count) =
+                if new_count <= MAX_COUNTED_WORDS {
-                    new_count.filter(|new_count| *new_count <= MAX_COUNTED_WORDS)
-                {
                     buffer.clear();
                     buffer.extend_from_slice(&fid.to_be_bytes());
                     buffer.push(new_count as u8);
@@ -239,15 +235,25 @@ impl<'a, 'extractor> Extractor<'extractor> for WordDocidsExtractorData<'a> {
 pub struct WordDocidsExtractors;
 
 impl WordDocidsExtractors {
-    pub fn run_extraction<'pl, 'fid, 'indexer, 'index, 'extractor, DC: DocumentChanges<'pl>, MSP>(
+    pub fn run_extraction<
+        'pl,
+        'fid,
+        'indexer,
+        'index,
+        'extractor,
+        DC: DocumentChanges<'pl>,
+        MSP,
+        SP,
+    >(
         grenad_parameters: GrenadParameters,
         document_changes: &DC,
-        indexing_context: IndexingContext<'fid, 'indexer, 'index, MSP>,
+        indexing_context: IndexingContext<'fid, 'indexer, 'index, MSP, SP>,
         extractor_allocs: &'extractor mut ThreadLocal<FullySend<Bump>>,
-        step: IndexingStep,
+        step: Step,
     ) -> Result<WordDocidsCaches<'extractor>>
     where
         MSP: Fn() -> bool + Sync,
+        SP: Fn(Progress) + Sync,
     {
         let index = indexing_context.index;
         let rtxn = index.read_txn()?;
@@ -345,15 +351,6 @@ impl WordDocidsExtractors {
                 )?;
             }
             DocumentChange::Update(inner) => {
-                if !inner.has_changed_for_fields(
-                    document_tokenizer.attribute_to_extract,
-                    &context.rtxn,
-                    context.index,
-                    context.db_fields_ids_map,
-                )? {
-                    return Ok(());
-                }
-
                 let mut token_fn = |fname: &str, fid, pos, word: &str| {
                     cached_sorter.insert_del_u32(
                         fid,

crates/milli/src/update/new/extract/searchable/extract_word_pair_proximity_docids.rs

@@ -70,15 +70,6 @@ impl SearchableExtractor for WordPairProximityDocidsExtractor {
                 )?;
             }
             DocumentChange::Update(inner) => {
-                if !inner.has_changed_for_fields(
-                    document_tokenizer.attribute_to_extract,
-                    rtxn,
-                    index,
-                    context.db_fields_ids_map,
-                )? {
-                    return Ok(());
-                }
-
                 let document = inner.current(rtxn, index, context.db_fields_ids_map)?;
                 process_document_tokens(
                     document,

crates/milli/src/update/new/extract/searchable/mod.rs

@@ -14,9 +14,9 @@ use tokenize_document::{tokenizer_builder, DocumentTokenizer};
 use super::cache::BalancedCaches;
 use super::DocidsExtractor;
 use crate::update::new::indexer::document_changes::{
-    extract, DocumentChangeContext, DocumentChanges, Extractor, IndexingContext,
+    extract, DocumentChangeContext, DocumentChanges, Extractor, IndexingContext, Progress,
 };
-use crate::update::new::steps::IndexingStep;
+use crate::update::new::steps::Step;
 use crate::update::new::thread_local::{FullySend, ThreadLocal};
 use crate::update::new::DocumentChange;
 use crate::update::GrenadParameters;
@@ -56,15 +56,16 @@ impl<'a, 'extractor, EX: SearchableExtractor + Sync> Extractor<'extractor>
 }
 
 pub trait SearchableExtractor: Sized + Sync {
-    fn run_extraction<'pl, 'fid, 'indexer, 'index, 'extractor, DC: DocumentChanges<'pl>, MSP>(
+    fn run_extraction<'pl, 'fid, 'indexer, 'index, 'extractor, DC: DocumentChanges<'pl>, MSP, SP>(
         grenad_parameters: GrenadParameters,
         document_changes: &DC,
-        indexing_context: IndexingContext<'fid, 'indexer, 'index, MSP>,
+        indexing_context: IndexingContext<'fid, 'indexer, 'index, MSP, SP>,
         extractor_allocs: &'extractor mut ThreadLocal<FullySend<Bump>>,
-        step: IndexingStep,
+        step: Step,
     ) -> Result<Vec<BalancedCaches<'extractor>>>
     where
         MSP: Fn() -> bool + Sync,
+        SP: Fn(Progress) + Sync,
     {
         let rtxn = indexing_context.index.read_txn()?;
         let stop_words = indexing_context.index.stop_words(&rtxn)?;
@@ -133,15 +134,16 @@ pub trait SearchableExtractor: Sized + Sync {
 }
 
 impl<T: SearchableExtractor> DocidsExtractor for T {
-    fn run_extraction<'pl, 'fid, 'indexer, 'index, 'extractor, DC: DocumentChanges<'pl>, MSP>(
+    fn run_extraction<'pl, 'fid, 'indexer, 'index, 'extractor, DC: DocumentChanges<'pl>, MSP, SP>(
         grenad_parameters: GrenadParameters,
         document_changes: &DC,
-        indexing_context: IndexingContext<'fid, 'indexer, 'index, MSP>,
+        indexing_context: IndexingContext<'fid, 'indexer, 'index, MSP, SP>,
         extractor_allocs: &'extractor mut ThreadLocal<FullySend<Bump>>,
-        step: IndexingStep,
+        step: Step,
     ) -> Result<Vec<BalancedCaches<'extractor>>>
     where
         MSP: Fn() -> bool + Sync,
+        SP: Fn(Progress) + Sync,
     {
         Self::run_extraction(
             grenad_parameters,

crates/milli/src/update/new/extract/searchable/tokenize_document.rs

@@ -176,10 +176,9 @@ pub fn tokenizer_builder<'a>(
 #[cfg(test)]
 mod test {
     use bumpalo::Bump;
-    use bumparaw_collections::RawMap;
     use charabia::TokenizerBuilder;
     use meili_snap::snapshot;
-    use rustc_hash::FxBuildHasher;
+    use raw_collections::RawMap;
     use serde_json::json;
     use serde_json::value::RawValue;
 
@@ -235,7 +234,7 @@ mod test {
 
         let bump = Bump::new();
         let document: &RawValue = serde_json::from_str(&document).unwrap();
-        let document = RawMap::from_raw_value_and_hasher(document, FxBuildHasher, &bump).unwrap();
+        let document = RawMap::from_raw_value(document, &bump).unwrap();
 
         let document = Versions::single(document);
         let document = DocumentFromVersions::new(&document);

crates/milli/src/update/new/extract/vectors/mod.rs

@@ -130,7 +130,6 @@ impl<'a, 'b, 'extractor> Extractor<'extractor> for EmbeddingExtractor<'a, 'b> {
                                 );
                             } else if new_vectors.regenerate {
                                 let new_rendered = prompt.render_document(
-                                    update.external_document_id(),
                                     update.current(
                                         &context.rtxn,
                                         context.index,
@@ -140,7 +139,6 @@ impl<'a, 'b, 'extractor> Extractor<'extractor> for EmbeddingExtractor<'a, 'b> {
                                     &context.doc_alloc,
                                 )?;
                                 let old_rendered = prompt.render_document(
-                                    update.external_document_id(),
                                     update.merged(
                                         &context.rtxn,
                                         context.index,
@@ -160,7 +158,6 @@ impl<'a, 'b, 'extractor> Extractor<'extractor> for EmbeddingExtractor<'a, 'b> {
                             }
                         } else if old_vectors.regenerate {
                             let old_rendered = prompt.render_document(
-                                update.external_document_id(),
                                 update.current(
                                     &context.rtxn,
                                     context.index,
@@ -170,7 +167,6 @@ impl<'a, 'b, 'extractor> Extractor<'extractor> for EmbeddingExtractor<'a, 'b> {
                                 &context.doc_alloc,
                             )?;
                             let new_rendered = prompt.render_document(
-                                update.external_document_id(),
                                 update.merged(
                                     &context.rtxn,
                                     context.index,
@@ -220,7 +216,6 @@ impl<'a, 'b, 'extractor> Extractor<'extractor> for EmbeddingExtractor<'a, 'b> {
                                 );
                             } else if new_vectors.regenerate {
                                 let rendered = prompt.render_document(
-                                    insertion.external_document_id(),
                                     insertion.inserted(),
                                     context.new_fields_ids_map,
                                     &context.doc_alloc,
@@ -234,7 +229,6 @@ impl<'a, 'b, 'extractor> Extractor<'extractor> for EmbeddingExtractor<'a, 'b> {
                             }
                         } else {
                             let rendered = prompt.render_document(
-                                insertion.external_document_id(),
                                 insertion.inserted(),
                                 context.new_fields_ids_map,
                                 &context.doc_alloc,

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

@@ -103,8 +103,6 @@ impl<'indexer> FacetSearchBuilder<'indexer> {
 
     #[tracing::instrument(level = "trace", skip_all, target = "indexing::facet_fst")]
     pub fn merge_and_write(self, index: &Index, wtxn: &mut RwTxn, rtxn: &RoTxn) -> Result<()> {
-        tracing::trace!("merge facet strings for facet search: {:?}", self.registered_facets);
-
         let reader = self.normalized_facet_string_docids_sorter.into_reader_cursors()?;
         let mut builder = grenad::MergerBuilder::new(MergeDeladdBtreesetString);
         builder.extend(reader);
@@ -120,15 +118,12 @@ impl<'indexer> FacetSearchBuilder<'indexer> {
                 BEU16StrCodec::bytes_decode(key).map_err(heed::Error::Encoding)?;
 
             if current_field_id != Some(field_id) {
-                if let (Some(current_field_id), Some(fst_merger_builder)) =
+                if let Some(fst_merger_builder) = fst_merger_builder {
-                    (current_field_id, fst_merger_builder)
-                {
                     let mmap = fst_merger_builder.build(&mut callback)?;
-                    index.facet_id_string_fst.remap_data_type::<Bytes>().put(
+                    index
-                        wtxn,
+                        .facet_id_string_fst
-                        &current_field_id,
+                        .remap_data_type::<Bytes>()
-                        &mmap,
+                        .put(wtxn, &field_id, &mmap)?;
-                    )?;
                 }
 
                 fst = index.facet_id_string_fst.get(rtxn, &field_id)?;

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

@@ -1,8 +1,6 @@
 use std::ops::ControlFlow;
 
 use bumpalo::Bump;
-use bumparaw_collections::RawVec;
-use rustc_hash::FxBuildHasher;
 use serde::de::{DeserializeSeed, Deserializer as _, Visitor};
 use serde_json::value::RawValue;
 
@@ -362,7 +360,7 @@ impl<'a> DeserrRawValue<'a> {
 }
 
 pub struct DeserrRawVec<'a> {
-    vec: RawVec<'a>,
+    vec: raw_collections::RawVec<'a>,
     alloc: &'a Bump,
 }
 
@@ -381,7 +379,7 @@ impl<'a> deserr::Sequence for DeserrRawVec<'a> {
 }
 
 pub struct DeserrRawVecIter<'a> {
-    it: bumparaw_collections::vec::iter::IntoIter<'a>,
+    it: raw_collections::vec::iter::IntoIter<'a>,
     alloc: &'a Bump,
 }
 
@@ -395,7 +393,7 @@ impl<'a> Iterator for DeserrRawVecIter<'a> {
 }
 
 pub struct DeserrRawMap<'a> {
-    map: bumparaw_collections::RawMap<'a, FxBuildHasher>,
+    map: raw_collections::RawMap<'a>,
     alloc: &'a Bump,
 }
 
@@ -418,7 +416,7 @@ impl<'a> deserr::Map for DeserrRawMap<'a> {
 }
 
 pub struct DeserrRawMapIter<'a> {
-    it: bumparaw_collections::map::iter::IntoIter<'a>,
+    it: raw_collections::map::iter::IntoIter<'a>,
     alloc: &'a Bump,
 }
 
@@ -617,7 +615,7 @@ impl<'de> Visitor<'de> for DeserrRawValueVisitor<'de> {
     where
         A: serde::de::SeqAccess<'de>,
     {
-        let mut raw_vec = RawVec::new_in(self.alloc);
+        let mut raw_vec = raw_collections::RawVec::new_in(self.alloc);
         while let Some(next) = seq.next_element()? {
             raw_vec.push(next);
         }