Fix the legend

3842: fix some typos r=dureuill a=cuishuang

# Pull Request

## Related issue
Fixes #<issue_number>

## What does this PR do?
- fix some typos

## PR checklist
Please check if your PR fulfills the following requirements:
- [x] Does this PR fix an existing issue, or have you listed the changes applied in the PR description (and why they are needed)?
- [x] Have you read the contributing guidelines?
- [x] Have you made sure that the title is accurate and descriptive of the changes?

Thank you so much for contributing to Meilisearch!


Co-authored-by: cui fliter <imcusg@gmail.com>

.github/scripts/check-release.sh

@@ -1,24 +1,41 @@
-#!/bin/bash
+#!/usr/bin/env bash
+set -eu -o pipefail
 
-# check_tag $current_tag $file_tag $file_name
-function check_tag {
-  if [[ "$1" != "$2" ]]; then
-      echo "Error: the current tag does not match the version in Cargo.toml: found $2 - expected $1"
-      ret=1
-  fi
+check_tag() {
+    local expected=$1
+    local actual=$2
+    local filename=$3
+
+    if [[ $actual != $expected ]]; then
+        echo >&2 "Error: the current tag does not match the version in $filename: found $actual, expected $expected"
+        return 1
+    fi
 }
 
+read_version() {
+    grep '^version = ' | cut -d \" -f 2
+}
+
+if [[ -z "${GITHUB_REF:-}" ]]; then
+    echo >&2 "Error: GITHUB_REF is not set"
+    exit 1
+fi
+
+if [[ ! "$GITHUB_REF" =~ ^refs/tags/v[0-9]+\.[0-9]+\.[0-9]+(-[a-z0-9]+)?$ ]]; then
+    echo >&2 "Error: GITHUB_REF is not a valid tag: $GITHUB_REF"
+    exit 1
+fi
+
+current_tag=${GITHUB_REF#refs/tags/v}
 ret=0
-current_tag=${GITHUB_REF#'refs/tags/v'}
 
-file_tag="$(grep '^version = ' Cargo.toml | cut -d '=' -f 2 | tr -d '"' | tr -d ' ')"
-check_tag $current_tag $file_tag
+toml_tag="$(cat Cargo.toml | read_version)"
+check_tag "$current_tag" "$toml_tag" Cargo.toml || ret=1
 
-lock_file='Cargo.lock'
-lock_tag=$(grep -A 1 'name = "meilisearch-auth"' $lock_file | grep version | cut -d '=' -f 2 | tr -d '"' | tr -d ' ')
-check_tag $current_tag $lock_tag $lock_file
+lock_tag=$(grep -A 1 '^name = "meilisearch-auth"' Cargo.lock | read_version)
+check_tag "$current_tag" "$lock_tag" Cargo.lock || ret=1
 
-if [[ "$ret" -eq 0 ]] ; then
-  echo 'OK'
+if (( ret == 0 )); then
+    echo 'OK'
 fi
 exit $ret

.github/workflows/fuzzer-indexing.yml

@@ -0,0 +1,24 @@
+name: Run the indexing fuzzer
+
+on:
+  push:
+    branches:
+      - main
+
+jobs:
+  fuzz:
+    name: Setup the action
+    runs-on: ubuntu-latest
+    timeout-minutes: 4320 # 72h
+    steps:
+      - uses: actions/checkout@v3
+      - uses: actions-rs/toolchain@v1
+        with:
+          profile: minimal
+          toolchain: stable
+          override: true
+
+      # Run benchmarks
+      - name: Run the fuzzer
+        run: |
+          cargo run --release --bin fuzz-indexing

.github/workflows/sdks-tests.yml

@@ -16,13 +16,28 @@ env:
   MEILI_NO_ANALYTICS: 'true'
 
 jobs:
+  define-docker-image:
+    runs-on: ubuntu-latest
+    outputs:
+      docker-image: ${{ steps.define-image.outputs.docker-image }}
+    steps:
+      - uses: actions/checkout@v3
+      - name: Define the Docker image we need to use
+        id: define-image
+        run: |
+          event=${{ github.event_name }}
+          echo "docker-image=nightly" >> $GITHUB_OUTPUT
+          if [[ $event == 'workflow_dispatch' ]]; then
+            echo "docker-image=${{ github.event.inputs.docker_image }}" >> $GITHUB_OUTPUT
+          fi
 
   meilisearch-js-tests:
+    needs: define-docker-image
     name: JS SDK tests
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ github.event.inputs.docker_image }}
+        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -52,11 +67,12 @@ jobs:
         run: yarn test:env:browser
 
   instant-meilisearch-tests:
+    needs: define-docker-image
     name: instant-meilisearch tests
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ github.event.inputs.docker_image }}
+        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -78,11 +94,12 @@ jobs:
         run: yarn build
 
   meilisearch-php-tests:
+    needs: define-docker-image
     name: PHP SDK tests
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ github.event.inputs.docker_image }}
+        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -108,11 +125,12 @@ jobs:
           composer remove --dev guzzlehttp/guzzle http-interop/http-factory-guzzle
 
   meilisearch-python-tests:
+    needs: define-docker-image
     name: Python SDK tests
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ github.event.inputs.docker_image }}
+        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -132,11 +150,12 @@ jobs:
         run: pipenv run pytest
 
   meilisearch-go-tests:
+    needs: define-docker-image
     name: Go SDK tests
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ github.event.inputs.docker_image }}
+        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -161,11 +180,12 @@ jobs:
         run: go test -v ./...
 
   meilisearch-ruby-tests:
+    needs: define-docker-image
     name: Ruby SDK tests
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ github.event.inputs.docker_image }}
+        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}
@@ -185,11 +205,12 @@ jobs:
         run: bundle exec rspec
 
   meilisearch-rust-tests:
+    needs: define-docker-image
     name: Rust SDK tests
     runs-on: ubuntu-latest
     services:
       meilisearch:
-        image: getmeili/meilisearch:${{ github.event.inputs.docker_image }}
+        image: getmeili/meilisearch:${{ needs.define-docker-image.outputs.docker-image }}
         env:
           MEILI_MASTER_KEY: ${{ env.MEILI_MASTER_KEY }}
           MEILI_NO_ANALYTICS: ${{ env.MEILI_NO_ANALYTICS }}

Cargo.toml

@@ -10,10 +10,12 @@ members = [
     "file-store",
     "permissive-json-pointer",
     "milli",
+    "index-stats",
     "filter-parser",
     "flatten-serde-json",
     "json-depth-checker",
-    "benchmarks"
+    "benchmarks",
+    "fuzzers",
 ]
 
 [workspace.package]

fuzzers/Cargo.toml

@@ -0,0 +1,20 @@
+[package]
+name = "fuzzers"
+publish = false
+
+version.workspace = true
+authors.workspace = true
+description.workspace = true
+homepage.workspace = true
+readme.workspace = true
+edition.workspace = true
+license.workspace = true
+
+[dependencies]
+arbitrary = { version = "1.3.0", features = ["derive"] }
+clap = { version = "4.3.0", features = ["derive"] }
+fastrand = "1.9.0"
+milli = { path = "../milli" }
+serde = { version = "1.0.160", features = ["derive"] }
+serde_json = { version = "1.0.95", features = ["preserve_order"] }
+tempfile = "3.5.0"

fuzzers/README.md

@@ -0,0 +1,3 @@
+# Fuzzers
+
+The purpose of this crate is to contains all the handmade "fuzzer" we may need.

fuzzers/src/bin/fuzz-indexing.rs

@@ -0,0 +1,152 @@
+use std::num::NonZeroUsize;
+use std::path::PathBuf;
+use std::sync::atomic::{AtomicBool, AtomicUsize, Ordering};
+use std::time::Duration;
+
+use arbitrary::{Arbitrary, Unstructured};
+use clap::Parser;
+use fuzzers::Operation;
+use milli::heed::EnvOpenOptions;
+use milli::update::{IndexDocuments, IndexDocumentsConfig, IndexerConfig};
+use milli::Index;
+use tempfile::TempDir;
+
+#[derive(Debug, Arbitrary)]
+struct Batch([Operation; 5]);
+
+#[derive(Debug, Clone, Parser)]
+struct Opt {
+    /// The number of fuzzer to run in parallel.
+    #[clap(long)]
+    par: Option<NonZeroUsize>,
+    // We need to put a lot of newlines in the following documentation or else everything gets collapsed on one line
+    /// The path in which the databases will be created.
+    /// Using a ramdisk is recommended.
+    ///
+    /// Linux:
+    ///
+    /// sudo mount -t tmpfs -o size=2g tmpfs ramdisk # to create it
+    ///
+    /// sudo umount ramdisk # to remove it
+    ///
+    /// MacOS:
+    ///
+    /// diskutil erasevolume HFS+ 'RAM Disk' `hdiutil attach -nobrowse -nomount ram://4194304 # create it
+    ///
+    /// hdiutil detach /dev/:the_disk
+    #[clap(long)]
+    path: Option<PathBuf>,
+}
+
+fn main() {
+    let opt = Opt::parse();
+    let progression: &'static AtomicUsize = Box::leak(Box::new(AtomicUsize::new(0)));
+    let stop: &'static AtomicBool = Box::leak(Box::new(AtomicBool::new(false)));
+
+    let par = opt.par.unwrap_or_else(|| std::thread::available_parallelism().unwrap()).get();
+    let mut handles = Vec::with_capacity(par);
+
+    for _ in 0..par {
+        let opt = opt.clone();
+
+        let handle = std::thread::spawn(move || {
+            let mut options = EnvOpenOptions::new();
+            options.map_size(1024 * 1024 * 1024 * 1024);
+            let tempdir = match opt.path {
+                Some(path) => TempDir::new_in(path).unwrap(),
+                None => TempDir::new().unwrap(),
+            };
+            let index = Index::new(options, tempdir.path()).unwrap();
+            let indexer_config = IndexerConfig::default();
+            let index_documents_config = IndexDocumentsConfig::default();
+
+            std::thread::scope(|s| {
+                loop {
+                    if stop.load(Ordering::Relaxed) {
+                        return;
+                    }
+                    let v: Vec<u8> =
+                        std::iter::repeat_with(|| fastrand::u8(..)).take(1000).collect();
+
+                    let mut data = Unstructured::new(&v);
+                    let batches = <[Batch; 5]>::arbitrary(&mut data).unwrap();
+                    // will be used to display the error once a thread crashes
+                    let dbg_input = format!("{:#?}", batches);
+
+                    let handle = s.spawn(|| {
+                        let mut wtxn = index.write_txn().unwrap();
+
+                        for batch in batches {
+                            let mut builder = IndexDocuments::new(
+                                &mut wtxn,
+                                &index,
+                                &indexer_config,
+                                index_documents_config.clone(),
+                                |_| (),
+                                || false,
+                            )
+                            .unwrap();
+
+                            for op in batch.0 {
+                                match op {
+                                    Operation::AddDoc(doc) => {
+                                        let documents =
+                                            milli::documents::objects_from_json_value(doc.to_d());
+                                        let documents =
+                                            milli::documents::documents_batch_reader_from_objects(
+                                                documents,
+                                            );
+                                        let (b, _added) = builder.add_documents(documents).unwrap();
+                                        builder = b;
+                                    }
+                                    Operation::DeleteDoc(id) => {
+                                        let (b, _removed) =
+                                            builder.remove_documents(vec![id.to_s()]).unwrap();
+                                        builder = b;
+                                    }
+                                }
+                            }
+                            builder.execute().unwrap();
+
+                            // after executing a batch we check if the database is corrupted
+                            let res = index.search(&wtxn).execute().unwrap();
+                            index.documents(&wtxn, res.documents_ids).unwrap();
+                            progression.fetch_add(1, Ordering::Relaxed);
+                        }
+                        wtxn.abort().unwrap();
+                    });
+                    if let err @ Err(_) = handle.join() {
+                        stop.store(true, Ordering::Relaxed);
+                        err.expect(&dbg_input);
+                    }
+                }
+            });
+        });
+        handles.push(handle);
+    }
+
+    std::thread::spawn(|| {
+        let mut last_value = 0;
+        let start = std::time::Instant::now();
+        loop {
+            let total = progression.load(Ordering::Relaxed);
+            let elapsed = start.elapsed().as_secs();
+            if elapsed > 3600 {
+                // after 1 hour, stop the fuzzer, success
+                std::process::exit(0);
+            }
+            println!(
+                "Has been running for {:?} seconds. Tested {} new values for a total of {}.",
+                elapsed,
+                total - last_value,
+                total
+            );
+            last_value = total;
+            std::thread::sleep(Duration::from_secs(1));
+        }
+    });
+
+    for handle in handles {
+        handle.join().unwrap();
+    }
+}

fuzzers/src/lib.rs

@@ -0,0 +1,46 @@
+use arbitrary::Arbitrary;
+use serde_json::{json, Value};
+
+#[derive(Debug, Arbitrary)]
+pub enum Document {
+    One,
+    Two,
+    Three,
+    Four,
+    Five,
+    Six,
+}
+
+impl Document {
+    pub fn to_d(&self) -> Value {
+        match self {
+            Document::One => json!({ "id": 0, "doggo": "bernese" }),
+            Document::Two => json!({ "id": 0, "doggo": "golden" }),
+            Document::Three => json!({ "id": 0, "catto": "jorts" }),
+            Document::Four => json!({ "id": 1, "doggo": "bernese" }),
+            Document::Five => json!({ "id": 1, "doggo": "golden" }),
+            Document::Six => json!({ "id": 1, "catto": "jorts" }),
+        }
+    }
+}
+
+#[derive(Debug, Arbitrary)]
+pub enum DocId {
+    Zero,
+    One,
+}
+
+impl DocId {
+    pub fn to_s(&self) -> String {
+        match self {
+            DocId::Zero => "0".to_string(),
+            DocId::One => "1".to_string(),
+        }
+    }
+}
+
+#[derive(Debug, Arbitrary)]
+pub enum Operation {
+    AddDoc(Document),
+    DeleteDoc(DocId),
+}

index-scheduler/src/autobatcher.rs

@@ -160,7 +160,7 @@ impl BatchKind {
 impl BatchKind {
     /// Returns a `ControlFlow::Break` if you must stop right now.
     /// The boolean tell you if an index has been created by the batched task.
-    /// To ease the writting of the code. `true` can be returned when you don't need to create an index
+    /// To ease the writing of the code. `true` can be returned when you don't need to create an index
     /// but false can't be returned if you needs to create an index.
     // TODO use an AutoBatchKind as input
     pub fn new(
@@ -214,7 +214,7 @@ impl BatchKind {
 
     /// Returns a `ControlFlow::Break` if you must stop right now.
     /// The boolean tell you if an index has been created by the batched task.
-    /// To ease the writting of the code. `true` can be returned when you don't need to create an index
+    /// To ease the writing of the code. `true` can be returned when you don't need to create an index
     /// but false can't be returned if you needs to create an index.
     #[rustfmt::skip]
     fn accumulate(self, id: TaskId, kind: AutobatchKind, index_already_exists: bool, primary_key: Option<&str>) -> ControlFlow<BatchKind, BatchKind> {
@@ -321,9 +321,18 @@ impl BatchKind {
                 })
             }
             (
-                this @ BatchKind::DocumentOperation { .. },
+                BatchKind::DocumentOperation { method, allow_index_creation, primary_key, mut operation_ids },
                 K::DocumentDeletion,
-            ) => Break(this),
+            ) => {
+                operation_ids.push(id);
+
+                Continue(BatchKind::DocumentOperation {
+                    method,
+                    allow_index_creation,
+                    primary_key,
+                    operation_ids,
+                })
+            }
             // but we can't autobatch documents if it's not the same kind
             // this match branch MUST be AFTER the previous one
             (
@@ -346,7 +355,35 @@ impl BatchKind {
                 deletion_ids.push(id);
                 Continue(BatchKind::DocumentClear { ids: deletion_ids })
             }
-            // we can't autobatch a deletion and an import
+            // we can autobatch the deletion and import if the index already exists
+            (
+                BatchKind::DocumentDeletion { mut deletion_ids },
+                K::DocumentImport { method, allow_index_creation, primary_key }
+            ) if index_already_exists => {
+                deletion_ids.push(id);
+
+                Continue(BatchKind::DocumentOperation {
+                    method,
+                    allow_index_creation,
+                    primary_key,
+                    operation_ids: deletion_ids,
+                })
+            }
+            // we can autobatch the deletion and import if both can't create an index
+            (
+                BatchKind::DocumentDeletion { mut deletion_ids },
+                K::DocumentImport { method, allow_index_creation, primary_key }
+            ) if !allow_index_creation => {
+                deletion_ids.push(id);
+
+                Continue(BatchKind::DocumentOperation {
+                    method,
+                    allow_index_creation,
+                    primary_key,
+                    operation_ids: deletion_ids,
+                })
+            }
+            // we can't autobatch a deletion and an import if the index does not exists but would be created by an addition
             (
                 this @ BatchKind::DocumentDeletion { .. },
                 K::DocumentImport { .. }
@@ -648,36 +685,36 @@ mod tests {
         debug_snapshot!(autobatch_from(false,None,  [settings(false)]), @"Some((Settings { allow_index_creation: false, settings_ids: [0] }, false))");
         debug_snapshot!(autobatch_from(false,None,  [settings(false), settings(false), settings(false)]), @"Some((Settings { allow_index_creation: false, settings_ids: [0, 1, 2] }, false))");
 
-        // We can't autobatch document addition with document deletion
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_del()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), doc_del()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), doc_del()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None), doc_del()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
-        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
-        debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, true, None), doc_del()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-        debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, true, None), doc_del()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0] }, true))");
-        debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, false, None), doc_del()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, false, None), doc_del()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
-        debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0] }, true))"###);
-        debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
-        debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0] }, false))"###);
-        // we also can't do the only way around
-        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, true, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, true, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, false, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, false, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(ReplaceDocuments, false, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
-        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(UpdateDocuments, false, Some("catto"))]), @"Some((DocumentDeletion { deletion_ids: [0] }, false))");
+        // We can autobatch document addition with document deletion
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, None), doc_del()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, None), doc_del()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, None), doc_del()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, None), doc_del()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, true))"###);
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, true))"###);
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(ReplaceDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+        debug_snapshot!(autobatch_from(true, None, [doc_imp(UpdateDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, true, None), doc_del()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, true, None), doc_del()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, true))");
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, false, None), doc_del()]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, false, None), doc_del()]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, true))"###);
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, true, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, true))"###);
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(ReplaceDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+        debug_snapshot!(autobatch_from(false, None, [doc_imp(UpdateDocuments, false, Some("catto")), doc_del()]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+        // And the other way around
+        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, true, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, true, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: None, operation_ids: [0, 1] }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, true, Some("catto"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, true, Some("catto"))]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: true, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(ReplaceDocuments, false, Some("catto"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+        debug_snapshot!(autobatch_from(true, None, [doc_del(), doc_imp(UpdateDocuments, false, Some("catto"))]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(ReplaceDocuments, false, None)]), @"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(UpdateDocuments, false, None)]), @"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: None, operation_ids: [0, 1] }, false))");
+        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(ReplaceDocuments, false, Some("catto"))]), @r###"Some((DocumentOperation { method: ReplaceDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
+        debug_snapshot!(autobatch_from(false, None, [doc_del(), doc_imp(UpdateDocuments, false, Some("catto"))]), @r###"Some((DocumentOperation { method: UpdateDocuments, allow_index_creation: false, primary_key: Some("catto"), operation_ids: [0, 1] }, false))"###);
     }
 
     #[test]

index-scheduler/src/batch.rs

@@ -998,7 +998,7 @@ impl IndexScheduler {
                 }()
                 .unwrap_or_default();
 
-                // The write transaction is directly owned and commited inside.
+                // The write transaction is directly owned and committed inside.
                 match self.index_mapper.delete_index(wtxn, &index_uid) {
                     Ok(()) => (),
                     Err(Error::IndexNotFound(_)) if index_has_been_created => (),

index-scheduler/src/lib.rs

@@ -1785,7 +1785,7 @@ mod tests {
             assert_eq!(task.kind.as_kind(), k);
         }
 
-        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "everything_is_succesfully_registered");
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "everything_is_successfully_registered");
     }
 
     #[test]
@@ -2075,6 +2075,105 @@ mod tests {
         snapshot!(snapshot_index_scheduler(&index_scheduler), name: "both_task_succeeded");
     }
 
+    #[test]
+    fn document_addition_and_document_deletion() {
+        let (index_scheduler, mut handle) = IndexScheduler::test(true, vec![]);
+
+        let content = r#"[
+            { "id": 1, "doggo": "jean bob" },
+            { "id": 2, "catto": "jorts" },
+            { "id": 3, "doggo": "bork" }
+        ]"#;
+
+        let (uuid, mut file) = index_scheduler.create_update_file_with_uuid(0).unwrap();
+        let documents_count = read_json(content.as_bytes(), file.as_file_mut()).unwrap();
+        file.persist().unwrap();
+        index_scheduler
+            .register(KindWithContent::DocumentAdditionOrUpdate {
+                index_uid: S("doggos"),
+                primary_key: Some(S("id")),
+                method: ReplaceDocuments,
+                content_file: uuid,
+                documents_count,
+                allow_index_creation: true,
+            })
+            .unwrap();
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "registered_the_first_task");
+        index_scheduler
+            .register(KindWithContent::DocumentDeletion {
+                index_uid: S("doggos"),
+                documents_ids: vec![S("1"), S("2")],
+            })
+            .unwrap();
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "registered_the_second_task");
+
+        handle.advance_one_successful_batch(); // The addition AND deletion should've been batched together
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "after_processing_the_batch");
+
+        let index = index_scheduler.index("doggos").unwrap();
+        let rtxn = index.read_txn().unwrap();
+        let field_ids_map = index.fields_ids_map(&rtxn).unwrap();
+        let field_ids = field_ids_map.ids().collect::<Vec<_>>();
+        let documents = index
+            .all_documents(&rtxn)
+            .unwrap()
+            .map(|ret| obkv_to_json(&field_ids, &field_ids_map, ret.unwrap().1).unwrap())
+            .collect::<Vec<_>>();
+        snapshot!(serde_json::to_string_pretty(&documents).unwrap(), name: "documents");
+    }
+
+    #[test]
+    fn document_deletion_and_document_addition() {
+        let (index_scheduler, mut handle) = IndexScheduler::test(true, vec![]);
+        index_scheduler
+            .register(KindWithContent::DocumentDeletion {
+                index_uid: S("doggos"),
+                documents_ids: vec![S("1"), S("2")],
+            })
+            .unwrap();
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "registered_the_first_task");
+
+        let content = r#"[
+            { "id": 1, "doggo": "jean bob" },
+            { "id": 2, "catto": "jorts" },
+            { "id": 3, "doggo": "bork" }
+        ]"#;
+
+        let (uuid, mut file) = index_scheduler.create_update_file_with_uuid(0).unwrap();
+        let documents_count = read_json(content.as_bytes(), file.as_file_mut()).unwrap();
+        file.persist().unwrap();
+        index_scheduler
+            .register(KindWithContent::DocumentAdditionOrUpdate {
+                index_uid: S("doggos"),
+                primary_key: Some(S("id")),
+                method: ReplaceDocuments,
+                content_file: uuid,
+                documents_count,
+                allow_index_creation: true,
+            })
+            .unwrap();
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "registered_the_second_task");
+
+        // The deletion should have failed because it can't create an index
+        handle.advance_one_failed_batch();
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "after_failing_the_deletion");
+
+        // The addition should works
+        handle.advance_one_successful_batch();
+        snapshot!(snapshot_index_scheduler(&index_scheduler), name: "after_last_successful_addition");
+
+        let index = index_scheduler.index("doggos").unwrap();
+        let rtxn = index.read_txn().unwrap();
+        let field_ids_map = index.fields_ids_map(&rtxn).unwrap();
+        let field_ids = field_ids_map.ids().collect::<Vec<_>>();
+        let documents = index
+            .all_documents(&rtxn)
+            .unwrap()
+            .map(|ret| obkv_to_json(&field_ids, &field_ids_map, ret.unwrap().1).unwrap())
+            .collect::<Vec<_>>();
+        snapshot!(serde_json::to_string_pretty(&documents).unwrap(), name: "documents");
+    }
+
     #[test]
     fn do_not_batch_task_of_different_indexes() {
         let (index_scheduler, mut handle) = IndexScheduler::test(true, vec![]);

index-scheduler/src/snapshots/lib.rs/document_addition_and_document_deletion/after_processing_the_batch.snap

@@ -0,0 +1,43 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, 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 }}
+1 {uid: 1, status: succeeded, details: { received_document_ids: 2, deleted_documents: Some(2) }, kind: DocumentDeletion { index_uid: "doggos", documents_ids: ["1", "2"] }}
+----------------------------------------------------------------------
+### Status:
+enqueued []
+succeeded [0,1,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [0,]
+"documentDeletion" [1,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,1,]
+----------------------------------------------------------------------
+### Index Mapper:
+doggos: { number_of_documents: 1, field_distribution: {"doggo": 1, "id": 1} }
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Started At:
+[timestamp] [0,1,]
+----------------------------------------------------------------------
+### Finished At:
+[timestamp] [0,1,]
+----------------------------------------------------------------------
+### File Store:
+
+----------------------------------------------------------------------
+

index-scheduler/src/snapshots/lib.rs/document_addition_and_document_deletion/documents.snap

@@ -0,0 +1,9 @@
+---
+source: index-scheduler/src/lib.rs
+---
+[
+  {
+    "id": 3,
+    "doggo": "bork"
+  }
+]

index-scheduler/src/snapshots/lib.rs/document_addition_and_document_deletion/registered_the_first_task.snap

@@ -0,0 +1,37 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: enqueued, details: { received_documents: 3, indexed_documents: None }, 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 }}
+----------------------------------------------------------------------
+### Status:
+enqueued [0,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,]
+----------------------------------------------------------------------
+### Index Mapper:
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+----------------------------------------------------------------------
+### Started At:
+----------------------------------------------------------------------
+### Finished At:
+----------------------------------------------------------------------
+### File Store:
+00000000-0000-0000-0000-000000000000
+
+----------------------------------------------------------------------
+

index-scheduler/src/snapshots/lib.rs/document_addition_and_document_deletion/registered_the_second_task.snap

@@ -0,0 +1,40 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: enqueued, details: { received_documents: 3, indexed_documents: None }, 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 }}
+1 {uid: 1, status: enqueued, details: { received_document_ids: 2, deleted_documents: None }, kind: DocumentDeletion { index_uid: "doggos", documents_ids: ["1", "2"] }}
+----------------------------------------------------------------------
+### Status:
+enqueued [0,1,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [0,]
+"documentDeletion" [1,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,1,]
+----------------------------------------------------------------------
+### Index Mapper:
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Started At:
+----------------------------------------------------------------------
+### Finished At:
+----------------------------------------------------------------------
+### File Store:
+00000000-0000-0000-0000-000000000000
+
+----------------------------------------------------------------------
+

index-scheduler/src/snapshots/lib.rs/document_deletion_and_document_addition/after_failing_the_deletion.snap

@@ -0,0 +1,43 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: failed, error: ResponseError { code: 200, message: "Index `doggos` not found.", error_code: "index_not_found", error_type: "invalid_request", error_link: "https://docs.meilisearch.com/errors#index_not_found" }, details: { received_document_ids: 2, deleted_documents: Some(0) }, kind: DocumentDeletion { index_uid: "doggos", documents_ids: ["1", "2"] }}
+1 {uid: 1, status: enqueued, details: { received_documents: 3, indexed_documents: None }, 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 }}
+----------------------------------------------------------------------
+### Status:
+enqueued [1,]
+failed [0,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [1,]
+"documentDeletion" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,1,]
+----------------------------------------------------------------------
+### Index Mapper:
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Started At:
+[timestamp] [0,]
+----------------------------------------------------------------------
+### Finished At:
+[timestamp] [0,]
+----------------------------------------------------------------------
+### File Store:
+00000000-0000-0000-0000-000000000000
+
+----------------------------------------------------------------------
+

index-scheduler/src/snapshots/lib.rs/document_deletion_and_document_addition/after_last_successful_addition.snap

@@ -0,0 +1,46 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: failed, error: ResponseError { code: 200, message: "Index `doggos` not found.", error_code: "index_not_found", error_type: "invalid_request", error_link: "https://docs.meilisearch.com/errors#index_not_found" }, details: { received_document_ids: 2, deleted_documents: Some(0) }, kind: DocumentDeletion { index_uid: "doggos", documents_ids: ["1", "2"] }}
+1 {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 }}
+----------------------------------------------------------------------
+### Status:
+enqueued []
+succeeded [1,]
+failed [0,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [1,]
+"documentDeletion" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,1,]
+----------------------------------------------------------------------
+### Index Mapper:
+doggos: { number_of_documents: 3, field_distribution: {"catto": 1, "doggo": 2, "id": 3} }
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Started At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Finished At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### File Store:
+
+----------------------------------------------------------------------
+

index-scheduler/src/snapshots/lib.rs/document_deletion_and_document_addition/documents.snap

@@ -0,0 +1,17 @@
+---
+source: index-scheduler/src/lib.rs
+---
+[
+  {
+    "id": 1,
+    "doggo": "jean bob"
+  },
+  {
+    "id": 2,
+    "catto": "jorts"
+  },
+  {
+    "id": 3,
+    "doggo": "bork"
+  }
+]

index-scheduler/src/snapshots/lib.rs/document_deletion_and_document_addition/registered_the_first_task.snap

@@ -0,0 +1,36 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: enqueued, details: { received_document_ids: 2, deleted_documents: None }, kind: DocumentDeletion { index_uid: "doggos", documents_ids: ["1", "2"] }}
+----------------------------------------------------------------------
+### Status:
+enqueued [0,]
+----------------------------------------------------------------------
+### Kind:
+"documentDeletion" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,]
+----------------------------------------------------------------------
+### Index Mapper:
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+----------------------------------------------------------------------
+### Started At:
+----------------------------------------------------------------------
+### Finished At:
+----------------------------------------------------------------------
+### File Store:
+
+----------------------------------------------------------------------
+

index-scheduler/src/snapshots/lib.rs/document_deletion_and_document_addition/registered_the_second_task.snap

@@ -0,0 +1,40 @@
+---
+source: index-scheduler/src/lib.rs
+---
+### Autobatching Enabled = true
+### Processing Tasks:
+[]
+----------------------------------------------------------------------
+### All Tasks:
+0 {uid: 0, status: enqueued, details: { received_document_ids: 2, deleted_documents: None }, kind: DocumentDeletion { index_uid: "doggos", documents_ids: ["1", "2"] }}
+1 {uid: 1, status: enqueued, details: { received_documents: 3, indexed_documents: None }, 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 }}
+----------------------------------------------------------------------
+### Status:
+enqueued [0,1,]
+----------------------------------------------------------------------
+### Kind:
+"documentAdditionOrUpdate" [1,]
+"documentDeletion" [0,]
+----------------------------------------------------------------------
+### Index Tasks:
+doggos [0,1,]
+----------------------------------------------------------------------
+### Index Mapper:
+
+----------------------------------------------------------------------
+### Canceled By:
+
+----------------------------------------------------------------------
+### Enqueued At:
+[timestamp] [0,]
+[timestamp] [1,]
+----------------------------------------------------------------------
+### Started At:
+----------------------------------------------------------------------
+### Finished At:
+----------------------------------------------------------------------
+### File Store:
+00000000-0000-0000-0000-000000000000
+
+----------------------------------------------------------------------
+

index-stats/Cargo.toml

@@ -0,0 +1,12 @@
+[package]
+name = "index-stats"
+description = "A small program that computes internal stats of a Meilisearch index"
+version = "0.1.0"
+edition = "2021"
+publish = false
+
+[dependencies]
+anyhow = "1.0.71"
+clap = { version = "4.3.5", features = ["derive"] }
+milli = { path = "../milli" }
+piechart = "1.0.0"

index-stats/src/main.rs

@@ -0,0 +1,224 @@
+use std::cmp::Reverse;
+use std::path::PathBuf;
+
+use clap::Parser;
+use milli::heed::{types::ByteSlice, EnvOpenOptions, PolyDatabase, RoTxn};
+use milli::index::db_name::*;
+use milli::index::Index;
+use piechart::{Chart, Color, Data};
+
+/// Simple program to greet a person
+#[derive(Parser, Debug)]
+#[command(author, version, about, long_about = None)]
+struct Args {
+    /// The path to the LMDB Meilisearch index database.
+    path: PathBuf,
+
+    /// The radius of the graphs
+    #[clap(long, default_value_t = 10)]
+    graph_radius: u16,
+
+    /// The radius of the graphs
+    #[clap(long, default_value_t = 6)]
+    graph_aspect_ratio: u16,
+}
+
+fn main() -> anyhow::Result<()> {
+    let Args { path, graph_radius, graph_aspect_ratio } = Args::parse();
+    let env = EnvOpenOptions::new().max_dbs(24).open(path)?;
+
+    // TODO not sure to keep that...
+    //      if removed put the pub(crate) back in the Index struct
+    matches!(
+        Option::<Index>::None,
+        Some(Index {
+            env: _,
+            main: _,
+            word_docids: _,
+            exact_word_docids: _,
+            word_prefix_docids: _,
+            exact_word_prefix_docids: _,
+            word_pair_proximity_docids: _,
+            word_prefix_pair_proximity_docids: _,
+            prefix_word_pair_proximity_docids: _,
+            word_position_docids: _,
+            word_fid_docids: _,
+            field_id_word_count_docids: _,
+            word_prefix_position_docids: _,
+            word_prefix_fid_docids: _,
+            script_language_docids: _,
+            facet_id_exists_docids: _,
+            facet_id_is_null_docids: _,
+            facet_id_is_empty_docids: _,
+            facet_id_f64_docids: _,
+            facet_id_string_docids: _,
+            field_id_docid_facet_f64s: _,
+            field_id_docid_facet_strings: _,
+            documents: _,
+        })
+    );
+
+    let mut wtxn = env.write_txn()?;
+    let main = env.create_poly_database(&mut wtxn, Some(MAIN))?;
+    let word_docids = env.create_poly_database(&mut wtxn, Some(WORD_DOCIDS))?;
+    let exact_word_docids = env.create_poly_database(&mut wtxn, Some(EXACT_WORD_DOCIDS))?;
+    let word_prefix_docids = env.create_poly_database(&mut wtxn, Some(WORD_PREFIX_DOCIDS))?;
+    let exact_word_prefix_docids =
+        env.create_poly_database(&mut wtxn, Some(EXACT_WORD_PREFIX_DOCIDS))?;
+    let word_pair_proximity_docids =
+        env.create_poly_database(&mut wtxn, Some(WORD_PAIR_PROXIMITY_DOCIDS))?;
+    let script_language_docids =
+        env.create_poly_database(&mut wtxn, Some(SCRIPT_LANGUAGE_DOCIDS))?;
+    let word_prefix_pair_proximity_docids =
+        env.create_poly_database(&mut wtxn, Some(WORD_PREFIX_PAIR_PROXIMITY_DOCIDS))?;
+    let prefix_word_pair_proximity_docids =
+        env.create_poly_database(&mut wtxn, Some(PREFIX_WORD_PAIR_PROXIMITY_DOCIDS))?;
+    let word_position_docids = env.create_poly_database(&mut wtxn, Some(WORD_POSITION_DOCIDS))?;
+    let word_fid_docids = env.create_poly_database(&mut wtxn, Some(WORD_FIELD_ID_DOCIDS))?;
+    let field_id_word_count_docids =
+        env.create_poly_database(&mut wtxn, Some(FIELD_ID_WORD_COUNT_DOCIDS))?;
+    let word_prefix_position_docids =
+        env.create_poly_database(&mut wtxn, Some(WORD_PREFIX_POSITION_DOCIDS))?;
+    let word_prefix_fid_docids =
+        env.create_poly_database(&mut wtxn, Some(WORD_PREFIX_FIELD_ID_DOCIDS))?;
+    let facet_id_f64_docids = env.create_poly_database(&mut wtxn, Some(FACET_ID_F64_DOCIDS))?;
+    let facet_id_string_docids =
+        env.create_poly_database(&mut wtxn, Some(FACET_ID_STRING_DOCIDS))?;
+    let facet_id_exists_docids =
+        env.create_poly_database(&mut wtxn, Some(FACET_ID_EXISTS_DOCIDS))?;
+    let facet_id_is_null_docids =
+        env.create_poly_database(&mut wtxn, Some(FACET_ID_IS_NULL_DOCIDS))?;
+    let facet_id_is_empty_docids =
+        env.create_poly_database(&mut wtxn, Some(FACET_ID_IS_EMPTY_DOCIDS))?;
+    let field_id_docid_facet_f64s =
+        env.create_poly_database(&mut wtxn, Some(FIELD_ID_DOCID_FACET_F64S))?;
+    let field_id_docid_facet_strings =
+        env.create_poly_database(&mut wtxn, Some(FIELD_ID_DOCID_FACET_STRINGS))?;
+    let documents = env.create_poly_database(&mut wtxn, Some(DOCUMENTS))?;
+    wtxn.commit()?;
+
+    let list = [
+        (main, MAIN),
+        (word_docids, WORD_DOCIDS),
+        (exact_word_docids, EXACT_WORD_DOCIDS),
+        (word_prefix_docids, WORD_PREFIX_DOCIDS),
+        (exact_word_prefix_docids, EXACT_WORD_PREFIX_DOCIDS),
+        (word_pair_proximity_docids, WORD_PAIR_PROXIMITY_DOCIDS),
+        (script_language_docids, SCRIPT_LANGUAGE_DOCIDS),
+        (word_prefix_pair_proximity_docids, WORD_PREFIX_PAIR_PROXIMITY_DOCIDS),
+        (prefix_word_pair_proximity_docids, PREFIX_WORD_PAIR_PROXIMITY_DOCIDS),
+        (word_position_docids, WORD_POSITION_DOCIDS),
+        (word_fid_docids, WORD_FIELD_ID_DOCIDS),
+        (field_id_word_count_docids, FIELD_ID_WORD_COUNT_DOCIDS),
+        (word_prefix_position_docids, WORD_PREFIX_POSITION_DOCIDS),
+        (word_prefix_fid_docids, WORD_PREFIX_FIELD_ID_DOCIDS),
+        (facet_id_f64_docids, FACET_ID_F64_DOCIDS),
+        (facet_id_string_docids, FACET_ID_STRING_DOCIDS),
+        (facet_id_exists_docids, FACET_ID_EXISTS_DOCIDS),
+        (facet_id_is_null_docids, FACET_ID_IS_NULL_DOCIDS),
+        (facet_id_is_empty_docids, FACET_ID_IS_EMPTY_DOCIDS),
+        (field_id_docid_facet_f64s, FIELD_ID_DOCID_FACET_F64S),
+        (field_id_docid_facet_strings, FIELD_ID_DOCID_FACET_STRINGS),
+        (documents, DOCUMENTS),
+    ];
+
+    let rtxn = env.read_txn()?;
+    let result: Result<Vec<_>, _> =
+        list.into_iter().map(|(db, name)| compute_stats(&rtxn, db).map(|s| (s, name))).collect();
+    let mut stats = result?;
+
+    println!("{:1$} Number of Entries", "", graph_radius as usize * 2);
+    stats.sort_by_key(|(s, _)| Reverse(s.number_of_entries));
+    let data = compute_graph_data(stats.iter().map(|(s, n)| (s.number_of_entries as f32, *n)));
+    Chart::new().radius(graph_radius).aspect_ratio(graph_aspect_ratio).draw(&data);
+    display_legend(&data);
+    print!("\r\n");
+
+    println!("{:1$} Size of Entries", "", graph_radius as usize * 2);
+    stats.sort_by_key(|(s, _)| Reverse(s.size_of_entries));
+    let data = compute_graph_data(stats.iter().map(|(s, n)| (s.size_of_entries as f32, *n)));
+    Chart::new().radius(graph_radius).aspect_ratio(graph_aspect_ratio).draw(&data);
+    display_legend(&data);
+    print!("\r\n");
+
+    println!("{:1$} Size of Data", "", graph_radius as usize * 2);
+    stats.sort_by_key(|(s, _)| Reverse(s.size_of_data));
+    let data = compute_graph_data(stats.iter().map(|(s, n)| (s.size_of_data as f32, *n)));
+    Chart::new().radius(graph_radius).aspect_ratio(graph_aspect_ratio).draw(&data);
+    display_legend(&data);
+    print!("\r\n");
+
+    println!("{:1$} Size of Keys", "", graph_radius as usize * 2);
+    stats.sort_by_key(|(s, _)| Reverse(s.size_of_keys));
+    let data = compute_graph_data(stats.iter().map(|(s, n)| (s.size_of_keys as f32, *n)));
+    Chart::new().radius(graph_radius).aspect_ratio(graph_aspect_ratio).draw(&data);
+    display_legend(&data);
+
+    Ok(())
+}
+
+fn display_legend(data: &[Data]) {
+    let total: f32 = data.iter().map(|d| d.value).sum();
+    for Data { label, value, color, fill } in data {
+        println!(
+            "{} {} {:.02}%",
+            color.unwrap().paint(fill.to_string()),
+            label,
+            value / total * 100.0
+        );
+    }
+}
+
+fn compute_graph_data<'a>(stats: impl IntoIterator<Item = (f32, &'a str)>) -> Vec<Data> {
+    let mut colors = [
+        Color::Red,
+        Color::Green,
+        Color::Yellow,
+        Color::Blue,
+        Color::Purple,
+        Color::Cyan,
+        Color::White,
+    ]
+    .into_iter()
+    .cycle();
+
+    let mut characters = ['▴', '▵', '▾', '▿', '▪', '▫', '•', '◦'].into_iter().cycle();
+
+    stats
+        .into_iter()
+        .map(|(value, name)| Data {
+            label: (*name).into(),
+            value,
+            color: Some(colors.next().unwrap().into()),
+            fill: characters.next().unwrap(),
+        })
+        .collect()
+}
+
+#[derive(Debug)]
+pub struct Stats {
+    pub number_of_entries: u64,
+    pub size_of_keys: u64,
+    pub size_of_data: u64,
+    pub size_of_entries: u64,
+}
+
+fn compute_stats(rtxn: &RoTxn, db: PolyDatabase) -> anyhow::Result<Stats> {
+    let mut number_of_entries = 0;
+    let mut size_of_keys = 0;
+    let mut size_of_data = 0;
+
+    for result in db.iter::<_, ByteSlice, ByteSlice>(rtxn)? {
+        let (key, data) = result?;
+        number_of_entries += 1;
+        size_of_keys += key.len() as u64;
+        size_of_data += data.len() as u64;
+    }
+
+    Ok(Stats {
+        number_of_entries,
+        size_of_keys,
+        size_of_data,
+        size_of_entries: size_of_keys + size_of_data,
+    })
+}

milli/Cargo.toml

@@ -75,9 +75,6 @@ maplit = "1.0.2"
 md5 = "0.7.0"
 rand = { version = "0.8.5", features = ["small_rng"] }
 
-[target.'cfg(fuzzing)'.dev-dependencies]
-fuzzcheck = "0.12.1"
-
 [features]
 all-tokenizations = ["charabia/default"]
 

milli/src/documents/mod.rs

@@ -111,7 +111,6 @@ pub enum Error {
     Io(#[from] io::Error),
 }
 
-#[cfg(test)]
 pub fn objects_from_json_value(json: serde_json::Value) -> Vec<crate::Object> {
     let documents = match json {
         object @ serde_json::Value::Object(_) => vec![object],
@@ -141,7 +140,6 @@ macro_rules! documents {
     }};
 }
 
-#[cfg(test)]
 pub fn documents_batch_reader_from_objects(
     objects: impl IntoIterator<Item = Object>,
 ) -> DocumentsBatchReader<std::io::Cursor<Vec<u8>>> {

milli/src/external_documents_ids.rs

@@ -106,22 +106,30 @@ impl<'a> ExternalDocumentsIds<'a> {
         map
     }
 
+    /// Return an fst of the combined hard and soft deleted ID.
+    pub fn to_fst<'b>(&'b self) -> fst::Result<Cow<'b, fst::Map<Cow<'a, [u8]>>>> {
+        if self.soft.is_empty() {
+            return Ok(Cow::Borrowed(&self.hard));
+        }
+        let union_op = self.hard.op().add(&self.soft).r#union();
+
+        let mut iter = union_op.into_stream();
+        let mut new_hard_builder = fst::MapBuilder::memory();
+        while let Some((external_id, marked_docids)) = iter.next() {
+            let value = indexed_last_value(marked_docids).unwrap();
+            if value != DELETED_ID {
+                new_hard_builder.insert(external_id, value)?;
+            }
+        }
+
+        drop(iter);
+
+        Ok(Cow::Owned(new_hard_builder.into_map().map_data(Cow::Owned)?))
+    }
+
     fn merge_soft_into_hard(&mut self) -> fst::Result<()> {
         if self.soft.len() >= self.hard.len() / 2 {
-            let union_op = self.hard.op().add(&self.soft).r#union();
-
-            let mut iter = union_op.into_stream();
-            let mut new_hard_builder = fst::MapBuilder::memory();
-            while let Some((external_id, marked_docids)) = iter.next() {
-                let value = indexed_last_value(marked_docids).unwrap();
-                if value != DELETED_ID {
-                    new_hard_builder.insert(external_id, value)?;
-                }
-            }
-
-            drop(iter);
-
-            self.hard = new_hard_builder.into_map().map_data(Cow::Owned)?;
+            self.hard = self.to_fst()?.into_owned();
             self.soft = fst::Map::default().map_data(Cow::Owned)?;
         }
 

milli/src/index.rs

@@ -93,10 +93,10 @@ pub mod db_name {
 #[derive(Clone)]
 pub struct Index {
     /// The LMDB environment which this index is associated with.
-    pub(crate) env: heed::Env,
+    pub env: heed::Env,
 
     /// Contains many different types (e.g. the fields ids map).
-    pub(crate) main: PolyDatabase,
+    pub main: PolyDatabase,
 
     /// A word and all the documents ids containing the word.
     pub word_docids: Database<Str, RoaringBitmapCodec>,
@@ -150,7 +150,7 @@ pub struct Index {
     pub field_id_docid_facet_strings: Database<FieldDocIdFacetStringCodec, Str>,
 
     /// Maps the document id to the document as an obkv store.
-    pub(crate) documents: Database<OwnedType<BEU32>, ObkvCodec>,
+    pub documents: Database<OwnedType<BEU32>, ObkvCodec>,
 }
 
 impl Index {
@@ -1466,9 +1466,9 @@ pub(crate) mod tests {
 
         db_snap!(index, field_distribution,
             @r###"
-        age              1     
-        id               2     
-        name             2     
+        age              1      |
+        id               2      |
+        name             2      |
         "###
         );
 
@@ -1486,9 +1486,9 @@ pub(crate) mod tests {
 
         db_snap!(index, field_distribution,
             @r###"
-        age              1     
-        id               2     
-        name             2     
+        age              1      |
+        id               2      |
+        name             2      |
         "###
         );
 
@@ -1502,9 +1502,9 @@ pub(crate) mod tests {
 
         db_snap!(index, field_distribution,
             @r###"
-        has_dog          1     
-        id               2     
-        name             2     
+        has_dog          1      |
+        id               2      |
+        name             2      |
         "###
         );
     }

milli/src/search/new/distinct.rs

@@ -26,7 +26,6 @@ pub fn apply_distinct_rule(
     ctx: &mut SearchContext,
     field_id: u16,
     candidates: &RoaringBitmap,
-    // TODO: add a universe here, such that the `excluded` are a subset of the universe?
 ) -> Result<DistinctOutput> {
     let mut excluded = RoaringBitmap::new();
     let mut remaining = RoaringBitmap::new();

milli/src/search/new/exact_attribute.rs

@@ -206,7 +206,7 @@ impl State {
             )?;
             intersection &= &candidates;
             if !intersection.is_empty() {
-                // TODO: although not really worth it in terms of performance,
+                // Although not really worth it in terms of performance,
                 // if would be good to put this in cache for the sake of consistency
                 let candidates_with_exact_word_count = if count_all_positions < u8::MAX as usize {
                     ctx.index

milli/src/search/new/interner.rs

@@ -32,7 +32,7 @@ impl<T> Interned<T> {
 #[derive(Clone)]
 pub struct DedupInterner<T> {
     stable_store: Vec<T>,
-    lookup: FxHashMap<T, Interned<T>>, // TODO: Arc
+    lookup: FxHashMap<T, Interned<T>>,
 }
 impl<T> Default for DedupInterner<T> {
     fn default() -> Self {

milli/src/search/new/limits.rs

@@ -1,5 +1,4 @@
 /// Maximum number of tokens we consider in a single search.
-// TODO: Loic, find proper value here so we don't overflow the interner.
 pub const MAX_TOKEN_COUNT: usize = 1_000;
 
 /// Maximum number of prefixes that can be derived from a single word.

milli/src/search/new/query_graph.rs

@@ -92,7 +92,7 @@ impl QueryGraph {
     /// which contains ngrams.
     pub fn from_query(
         ctx: &mut SearchContext,
-        // NOTE: the terms here must be consecutive
+        // The terms here must be consecutive
         terms: &[LocatedQueryTerm],
     ) -> Result<(QueryGraph, Vec<LocatedQueryTerm>)> {
         let mut new_located_query_terms = terms.to_vec();
@@ -103,7 +103,7 @@ impl QueryGraph {
         let root_node = 0;
         let end_node = 1;
 
-        // TODO: we could consider generalizing to 4,5,6,7,etc. ngrams
+        // Ee could consider generalizing to 4,5,6,7,etc. ngrams
         let (mut prev2, mut prev1, mut prev0): (Vec<u16>, Vec<u16>, Vec<u16>) =
             (vec![], vec![], vec![root_node]);
 

milli/src/search/new/query_term/mod.rs

@@ -132,7 +132,6 @@ impl QueryTermSubset {
         if full_query_term.ngram_words.is_some() {
             return None;
         }
-        // TODO: included in subset
         if let Some(phrase) = full_query_term.zero_typo.phrase {
             self.zero_typo_subset.contains_phrase(phrase).then_some(ExactTerm::Phrase(phrase))
         } else if let Some(word) = full_query_term.zero_typo.exact {
@@ -182,7 +181,6 @@ impl QueryTermSubset {
         let word = match &self.zero_typo_subset {
             NTypoTermSubset::All => Some(use_prefix_db),
             NTypoTermSubset::Subset { words, phrases: _ } => {
-                // TODO: use a subset of prefix words instead
                 if words.contains(&use_prefix_db) {
                     Some(use_prefix_db)
                 } else {
@@ -204,7 +202,6 @@ impl QueryTermSubset {
         ctx: &mut SearchContext,
     ) -> Result<BTreeSet<Word>> {
         let mut result = BTreeSet::default();
-        // TODO: a compute_partially funtion
         if !self.one_typo_subset.is_empty() || !self.two_typo_subset.is_empty() {
             self.original.compute_fully_if_needed(ctx)?;
         }
@@ -300,7 +297,6 @@ impl QueryTermSubset {
         let mut result = BTreeSet::default();
 
         if !self.one_typo_subset.is_empty() {
-            // TODO: compute less than fully if possible
             self.original.compute_fully_if_needed(ctx)?;
         }
         let original = ctx.term_interner.get_mut(self.original);

milli/src/search/new/query_term/parse_query.rs

@@ -139,7 +139,6 @@ pub fn number_of_typos_allowed<'ctx>(
     let min_len_one_typo = ctx.index.min_word_len_one_typo(ctx.txn)?;
     let min_len_two_typos = ctx.index.min_word_len_two_typos(ctx.txn)?;
 
-    // TODO: should `exact_words` also disable prefix search, ngrams, split words, or synonyms?
     let exact_words = ctx.index.exact_words(ctx.txn)?;
 
     Ok(Box::new(move |word: &str| {
@@ -250,8 +249,6 @@ impl PhraseBuilder {
         } else {
             // token has kind Word
             let word = ctx.word_interner.insert(token.lemma().to_string());
-            // TODO: in a phrase, check that every word exists
-            // otherwise return an empty term
             self.words.push(Some(word));
         }
     }

milli/src/search/new/ranking_rule_graph/cheapest_paths.rs

@@ -1,5 +1,48 @@
-#![allow(clippy::too_many_arguments)]
+/** Implements a "PathVisitor" which finds all paths of a certain cost
+from the START to END node of a ranking rule graph.
 
+A path is a list of conditions. A condition is the data associated with
+an edge, given by the ranking rule. Some edges don't have a condition associated
+with them, they are "unconditional". These kinds of edges are used to "skip" a node.
+
+The algorithm uses a depth-first search. It benefits from two main optimisations:
+- The list of all possible costs to go from any node to the END node is precomputed
+- The `DeadEndsCache` reduces the number of valid paths drastically, by making some edges
+untraversable depending on what other edges were selected.
+
+These two optimisations are meant to avoid traversing edges that wouldn't lead
+to a valid path. In practically all cases, we avoid the exponential complexity
+that is inherent to depth-first search in a large ranking rule graph.
+
+The DeadEndsCache is a sort of prefix tree which associates a list of forbidden
+conditions to a list of traversed conditions.
+For example, the DeadEndsCache could say the following:
+- Immediately, from the start, the conditions `[a,b]` are forbidden
+    - if we take the condition `c`, then the conditions `[e]` are also forbidden
+        - and if after that, we take `f`, then `[h,i]` are also forbidden
+            - etc.
+    - if we take `g`, then `[f]` is also forbidden
+        - etc.
+    - etc.
+As we traverse the graph, we also traverse the `DeadEndsCache` and keep a list of forbidden
+conditions in memory. Then, we know to avoid all edges which have a condition that is forbidden.
+
+When a path is found from START to END, we give it to the `visit` closure.
+This closure takes a mutable reference to the `DeadEndsCache`. This means that
+the caller can update this cache. Therefore, we must handle the case where the
+DeadEndsCache has been updated. This means potentially backtracking up to the point
+where the traversed conditions are all allowed by the new DeadEndsCache.
+
+The algorithm also implements the `TermsMatchingStrategy` logic.
+Some edges are augmented with a list of "nodes_to_skip". Skipping
+a node means "reaching this node through an unconditional edge". If we have
+already traversed (ie. not skipped) a node that is in this list, then we know that we
+can't traverse this edge. Otherwise, we traverse the edge but make sure to skip any
+future node that was present in the "nodes_to_skip" list.
+
+The caller can decide to stop the path finding algorithm
+by returning a `ControlFlow::Break` from the `visit` closure.
+*/
 use std::collections::{BTreeSet, VecDeque};
 use std::iter::FromIterator;
 use std::ops::ControlFlow;
@@ -12,30 +55,41 @@ use crate::search::new::query_graph::QueryNode;
 use crate::search::new::small_bitmap::SmallBitmap;
 use crate::Result;
 
+/// Closure which processes a path found by the `PathVisitor`
 type VisitFn<'f, G> = &'f mut dyn FnMut(
+    // the path as a list of conditions
     &[Interned<<G as RankingRuleGraphTrait>::Condition>],
     &mut RankingRuleGraph<G>,
+    // a mutable reference to the DeadEndsCache, to update it in case the given
+    // path doesn't resolve to any valid document ids
     &mut DeadEndsCache<<G as RankingRuleGraphTrait>::Condition>,
 ) -> Result<ControlFlow<()>>;
 
+/// A structure which is kept but not updated during the traversal of the graph.
+/// It can however be updated by the `visit` closure once a valid path has been found.
 struct VisitorContext<'a, G: RankingRuleGraphTrait> {
     graph: &'a mut RankingRuleGraph<G>,
     all_costs_from_node: &'a MappedInterner<QueryNode, Vec<u64>>,
     dead_ends_cache: &'a mut DeadEndsCache<G::Condition>,
 }
 
+/// The internal state of the traversal algorithm
 struct VisitorState<G: RankingRuleGraphTrait> {
+    /// Budget from the current node to the end node
     remaining_cost: u64,
-
+    /// Previously visited conditions, in order.
     path: Vec<Interned<G::Condition>>,
-
+    /// Previously visited conditions, as an efficient and compact set.
     visited_conditions: SmallBitmap<G::Condition>,
+    /// Previously visited (ie not skipped) nodes, as an efficient and compact set.
     visited_nodes: SmallBitmap<QueryNode>,
-
+    /// The conditions that cannot be visited anymore
     forbidden_conditions: SmallBitmap<G::Condition>,
-    forbidden_conditions_to_nodes: SmallBitmap<QueryNode>,
+    /// The nodes that cannot be visited anymore (they must be skipped)
+    nodes_to_skip: SmallBitmap<QueryNode>,
 }
 
+/// See module documentation
 pub struct PathVisitor<'a, G: RankingRuleGraphTrait> {
     state: VisitorState<G>,
     ctx: VisitorContext<'a, G>,
@@ -56,14 +110,13 @@ impl<'a, G: RankingRuleGraphTrait> PathVisitor<'a, G> {
                 forbidden_conditions: SmallBitmap::for_interned_values_in(
                     &graph.conditions_interner,
                 ),
-                forbidden_conditions_to_nodes: SmallBitmap::for_interned_values_in(
-                    &graph.query_graph.nodes,
-                ),
+                nodes_to_skip: SmallBitmap::for_interned_values_in(&graph.query_graph.nodes),
             },
             ctx: VisitorContext { graph, all_costs_from_node, dead_ends_cache },
         }
     }
 
+    /// See module documentation
     pub fn visit_paths(mut self, visit: VisitFn<G>) -> Result<()> {
         let _ =
             self.state.visit_node(self.ctx.graph.query_graph.root_node, visit, &mut self.ctx)?;
@@ -72,22 +125,31 @@ impl<'a, G: RankingRuleGraphTrait> PathVisitor<'a, G> {
 }
 
 impl<G: RankingRuleGraphTrait> VisitorState<G> {
+    /// Visits a node: traverse all its valid conditional and unconditional edges.
+    ///
+    /// Returns ControlFlow::Break if the path finding algorithm should stop.
+    /// Returns whether a valid path was found from this node otherwise.
     fn visit_node(
         &mut self,
         from_node: Interned<QueryNode>,
         visit: VisitFn<G>,
         ctx: &mut VisitorContext<G>,
     ) -> Result<ControlFlow<(), bool>> {
+        // any valid path will be found from this point
+        // if a valid path was found, then we know that the DeadEndsCache may have been updated,
+        // and we will need to do more work to potentially backtrack
         let mut any_valid = false;
 
         let edges = ctx.graph.edges_of_node.get(from_node).clone();
         for edge_idx in edges.iter() {
+            // could be none if the edge was deleted
             let Some(edge) = ctx.graph.edges_store.get(edge_idx).clone() else { continue };
 
             if self.remaining_cost < edge.cost as u64 {
                 continue;
             }
             self.remaining_cost -= edge.cost as u64;
+
             let cf = match edge.condition {
                 Some(condition) => self.visit_condition(
                     condition,
@@ -119,6 +181,10 @@ impl<G: RankingRuleGraphTrait> VisitorState<G> {
         Ok(ControlFlow::Continue(any_valid))
     }
 
+    /// Visits an unconditional edge.
+    ///
+    /// Returns ControlFlow::Break if the path finding algorithm should stop.
+    /// Returns whether a valid path was found from this node otherwise.
     fn visit_no_condition(
         &mut self,
         dest_node: Interned<QueryNode>,
@@ -134,20 +200,29 @@ impl<G: RankingRuleGraphTrait> VisitorState<G> {
         {
             return Ok(ControlFlow::Continue(false));
         }
+        // We've reached the END node!
         if dest_node == ctx.graph.query_graph.end_node {
             let control_flow = visit(&self.path, ctx.graph, ctx.dead_ends_cache)?;
+            // We could change the return type of the visit closure such that the caller
+            // tells us whether the dead ends cache was updated or not.
+            // Alternatively, maybe the DeadEndsCache should have a generation number
+            // to it, so that we don't need to play with these booleans at all.
             match control_flow {
                 ControlFlow::Continue(_) => Ok(ControlFlow::Continue(true)),
                 ControlFlow::Break(_) => Ok(ControlFlow::Break(())),
             }
         } else {
-            let old_fbct = self.forbidden_conditions_to_nodes.clone();
-            self.forbidden_conditions_to_nodes.union(edge_new_nodes_to_skip);
+            let old_fbct = self.nodes_to_skip.clone();
+            self.nodes_to_skip.union(edge_new_nodes_to_skip);
             let cf = self.visit_node(dest_node, visit, ctx)?;
-            self.forbidden_conditions_to_nodes = old_fbct;
+            self.nodes_to_skip = old_fbct;
             Ok(cf)
         }
     }
+    /// Visits a conditional edge.
+    ///
+    /// Returns ControlFlow::Break if the path finding algorithm should stop.
+    /// Returns whether a valid path was found from this node otherwise.
     fn visit_condition(
         &mut self,
         condition: Interned<G::Condition>,
@@ -159,7 +234,7 @@ impl<G: RankingRuleGraphTrait> VisitorState<G> {
         assert!(dest_node != ctx.graph.query_graph.end_node);
 
         if self.forbidden_conditions.contains(condition)
-            || self.forbidden_conditions_to_nodes.contains(dest_node)
+            || self.nodes_to_skip.contains(dest_node)
             || edge_new_nodes_to_skip.intersects(&self.visited_nodes)
         {
             return Ok(ControlFlow::Continue(false));
@@ -180,19 +255,19 @@ impl<G: RankingRuleGraphTrait> VisitorState<G> {
         self.visited_nodes.insert(dest_node);
         self.visited_conditions.insert(condition);
 
-        let old_fc = self.forbidden_conditions.clone();
+        let old_forb_cond = self.forbidden_conditions.clone();
         if let Some(next_forbidden) =
             ctx.dead_ends_cache.forbidden_conditions_after_prefix(self.path.iter().copied())
         {
             self.forbidden_conditions.union(&next_forbidden);
         }
-        let old_fctn = self.forbidden_conditions_to_nodes.clone();
-        self.forbidden_conditions_to_nodes.union(edge_new_nodes_to_skip);
+        let old_nodes_to_skip = self.nodes_to_skip.clone();
+        self.nodes_to_skip.union(edge_new_nodes_to_skip);
 
         let cf = self.visit_node(dest_node, visit, ctx)?;
 
-        self.forbidden_conditions_to_nodes = old_fctn;
-        self.forbidden_conditions = old_fc;
+        self.nodes_to_skip = old_nodes_to_skip;
+        self.forbidden_conditions = old_forb_cond;
 
         self.visited_conditions.remove(condition);
         self.visited_nodes.remove(dest_node);

milli/src/search/new/ranking_rule_graph/condition_docids_cache.rs

@@ -9,12 +9,8 @@ use crate::search::new::query_term::LocatedQueryTermSubset;
 use crate::search::new::SearchContext;
 use crate::Result;
 
-// TODO: give a generation to each universe, then be able to get the exact
-// delta of docids between two universes of different generations!
-
 /// A cache storing the document ids associated with each ranking rule edge
 pub struct ConditionDocIdsCache<G: RankingRuleGraphTrait> {
-    // TOOD: should be a mapped interner?
     pub cache: FxHashMap<Interned<G::Condition>, ComputedCondition>,
     _phantom: PhantomData<G>,
 }
@@ -54,7 +50,7 @@ impl<G: RankingRuleGraphTrait> ConditionDocIdsCache<G> {
         }
         let condition = graph.conditions_interner.get_mut(interned_condition);
         let computed = G::resolve_condition(ctx, condition, universe)?;
-        // TODO: if computed.universe_len != universe.len() ?
+        // Can we put an assert here for computed.universe_len == universe.len() ?
         let _ = self.cache.insert(interned_condition, computed);
         let computed = &self.cache[&interned_condition];
         Ok(computed)

milli/src/search/new/ranking_rule_graph/dead_ends_cache.rs

@@ -2,6 +2,7 @@ use crate::search::new::interner::{FixedSizeInterner, Interned};
 use crate::search::new::small_bitmap::SmallBitmap;
 
 pub struct DeadEndsCache<T> {
+    // conditions and next could/should be part of the same vector
     conditions: Vec<Interned<T>>,
     next: Vec<Self>,
     pub forbidden: SmallBitmap<T>,
@@ -27,7 +28,7 @@ impl<T> DeadEndsCache<T> {
         self.forbidden.insert(condition);
     }
 
-    pub fn advance(&mut self, condition: Interned<T>) -> Option<&mut Self> {
+    fn advance(&mut self, condition: Interned<T>) -> Option<&mut Self> {
         if let Some(idx) = self.conditions.iter().position(|c| *c == condition) {
             Some(&mut self.next[idx])
         } else {

milli/src/search/new/ranking_rule_graph/fid/mod.rs

@@ -69,14 +69,9 @@ impl RankingRuleGraphTrait for FidGraph {
 
         let mut edges = vec![];
         for fid in all_fields {
-            // TODO: We can improve performances and relevancy by storing
-            //       the term subsets associated to each field ids fetched.
             edges.push((
-                fid as u32 * term.term_ids.len() as u32, // TODO improve the fid score i.e. fid^10.
-                conditions_interner.insert(FidCondition {
-                    term: term.clone(), // TODO remove this ugly clone
-                    fid,
-                }),
+                fid as u32 * term.term_ids.len() as u32,
+                conditions_interner.insert(FidCondition { term: term.clone(), fid }),
             ));
         }
 

milli/src/search/new/ranking_rule_graph/position/mod.rs

@@ -94,14 +94,9 @@ impl RankingRuleGraphTrait for PositionGraph {
         let mut edges = vec![];
 
         for (cost, positions) in positions_for_costs {
-            // TODO: We can improve performances and relevancy by storing
-            //       the term subsets associated to each position fetched
             edges.push((
                 cost,
-                conditions_interner.insert(PositionCondition {
-                    term: term.clone(), // TODO remove this ugly clone
-                    positions,
-                }),
+                conditions_interner.insert(PositionCondition { term: term.clone(), positions }),
             ));
         }
 

milli/src/search/new/ranking_rule_graph/proximity/compute_docids.rs

@@ -65,13 +65,6 @@ pub fn compute_docids(
         }
     }
 
-    // TODO: add safeguard in case the cartesian product is too large!
-    // even if we restrict the word derivations to a maximum of 100, the size of the
-    // caterisan product could reach a maximum of 10_000 derivations, which is way too much.
-    // Maybe prioritise the product of zero typo derivations, then the product of zero-typo/one-typo
-    // + one-typo/zero-typo, then one-typo/one-typo, then ... until an arbitrary limit has been
-    // reached
-
     for (left_phrase, left_word) in last_words_of_term_derivations(ctx, &left_term.term_subset)? {
         // Before computing the edges, check that the left word and left phrase
         // aren't disjoint with the universe, but only do it if there is more than
@@ -111,8 +104,6 @@ pub fn compute_docids(
     Ok(ComputedCondition {
         docids,
         universe_len: universe.len(),
-        // TODO: think about whether we want to reduce the subset,
-        // we probably should!
         start_term_subset: Some(left_term.clone()),
         end_term_subset: right_term.clone(),
     })
@@ -203,12 +194,7 @@ fn compute_non_prefix_edges(
             *docids |= new_docids;
         }
     }
-    if backward_proximity >= 1
-            // TODO: for now, we don't do any swapping when either term is a phrase
-            // but maybe we should. We'd need to look at the first/last word of the phrase
-            // depending on the context.
-            && left_phrase.is_none() && right_phrase.is_none()
-    {
+    if backward_proximity >= 1 && left_phrase.is_none() && right_phrase.is_none() {
         if let Some(new_docids) =
             ctx.get_db_word_pair_proximity_docids(word2, word1, backward_proximity)?
         {

milli/src/search/new/resolve_query_graph.rs

@@ -33,8 +33,6 @@ pub fn compute_query_term_subset_docids(
     ctx: &mut SearchContext,
     term: &QueryTermSubset,
 ) -> Result<RoaringBitmap> {
-    // TODO Use the roaring::MultiOps trait
-
     let mut docids = RoaringBitmap::new();
     for word in term.all_single_words_except_prefix_db(ctx)? {
         if let Some(word_docids) = ctx.word_docids(word)? {
@@ -59,8 +57,6 @@ pub fn compute_query_term_subset_docids_within_field_id(
     term: &QueryTermSubset,
     fid: u16,
 ) -> Result<RoaringBitmap> {
-    // TODO Use the roaring::MultiOps trait
-
     let mut docids = RoaringBitmap::new();
     for word in term.all_single_words_except_prefix_db(ctx)? {
         if let Some(word_fid_docids) = ctx.get_db_word_fid_docids(word.interned(), fid)? {
@@ -71,7 +67,6 @@ pub fn compute_query_term_subset_docids_within_field_id(
     for phrase in term.all_phrases(ctx)? {
         // There may be false positives when resolving a phrase, so we're not
         // guaranteed that all of its words are within a single fid.
-        // TODO: fix this?
         if let Some(word) = phrase.words(ctx).iter().flatten().next() {
             if let Some(word_fid_docids) = ctx.get_db_word_fid_docids(*word, fid)? {
                 docids |= ctx.get_phrase_docids(phrase)? & word_fid_docids;
@@ -95,7 +90,6 @@ pub fn compute_query_term_subset_docids_within_position(
     term: &QueryTermSubset,
     position: u16,
 ) -> Result<RoaringBitmap> {
-    // TODO Use the roaring::MultiOps trait
     let mut docids = RoaringBitmap::new();
     for word in term.all_single_words_except_prefix_db(ctx)? {
         if let Some(word_position_docids) =
@@ -108,7 +102,6 @@ pub fn compute_query_term_subset_docids_within_position(
     for phrase in term.all_phrases(ctx)? {
         // It's difficult to know the expected position of the words in the phrase,
         // so instead we just check the first one.
-        // TODO: fix this?
         if let Some(word) = phrase.words(ctx).iter().flatten().next() {
             if let Some(word_position_docids) = ctx.get_db_word_position_docids(*word, position)? {
                 docids |= ctx.get_phrase_docids(phrase)? & word_position_docids
@@ -132,9 +125,6 @@ pub fn compute_query_graph_docids(
     q: &QueryGraph,
     universe: &RoaringBitmap,
 ) -> Result<RoaringBitmap> {
-    // TODO: there must be a faster way to compute this big
-    // roaring bitmap expression
-
     let mut nodes_resolved = SmallBitmap::for_interned_values_in(&q.nodes);
     let mut path_nodes_docids = q.nodes.map(|_| RoaringBitmap::new());
 

milli/src/search/new/sort.rs

@@ -141,10 +141,6 @@ impl<'ctx, Query: RankingRuleQueryTrait> RankingRule<'ctx, Query> for Sort<'ctx,
         universe: &RoaringBitmap,
     ) -> Result<Option<RankingRuleOutput<Query>>> {
         let iter = self.iter.as_mut().unwrap();
-        // TODO: we should make use of the universe in the function below
-        // good for correctness, but ideally iter.next_bucket would take the current universe into account,
-        // as right now it could return buckets that don't intersect with the universe, meaning we will make many
-        // unneeded calls.
         if let Some(mut bucket) = iter.next_bucket()? {
             bucket.candidates &= universe;
             Ok(Some(bucket))

milli/src/search/new/tests/distinct.rs

@@ -527,7 +527,7 @@ fn test_distinct_all_candidates() {
     let SearchResult { documents_ids, candidates, .. } = s.execute().unwrap();
     let candidates = candidates.iter().collect::<Vec<_>>();
     insta::assert_snapshot!(format!("{documents_ids:?}"), @"[14, 26, 4, 7, 17, 23, 1, 19, 25, 8, 20, 24]");
-    // TODO: this is incorrect!
+    // This is incorrect, but unfortunately impossible to do better efficiently.
     insta::assert_snapshot!(format!("{candidates:?}"), @"[1, 4, 7, 8, 14, 17, 19, 20, 23, 24, 25, 26]");
 }
 

milli/src/search/new/tests/proximity.rs

@@ -122,11 +122,11 @@ fn create_edge_cases_index() -> TempIndex {
             sta stb stc ste stf stg sth sti stj stk stl stm stn sto stp stq str stst stt stu stv stw stx sty stz
             "
         },
-        // The next 5 documents lay out a trap with the split word, phrase search, or synonym `sun flower`. 
-        // If the search query is "sunflower", the split word "Sun Flower" will match some documents. 
+        // The next 5 documents lay out a trap with the split word, phrase search, or synonym `sun flower`.
+        // If the search query is "sunflower", the split word "Sun Flower" will match some documents.
         // If the query is `sunflower wilting`, then we should make sure that
-        // the sprximity condition `flower wilting: sprx N` also comes with the condition
-        // `sun wilting: sprx N+1`. TODO: this is not the exact condition we use for now. 
+        // the proximity condition `flower wilting: sprx N` also comes with the condition
+        // `sun wilting: sprx N+1`, but this is not the exact condition we use for now.
         // We only check that the phrase `sun flower` exists and `flower wilting: sprx N`, which
         // is better than nothing but not the best.
         {
@@ -139,7 +139,7 @@ fn create_edge_cases_index() -> TempIndex {
         },
         {
             "id": 3,
-            // This document matches the query `sunflower wilting`, but the sprximity condition 
+            // This document matches the query `sunflower wilting`, but the sprximity condition
             // between `sunflower` and `wilting` cannot be through the split-word `Sun Flower`
             // which would reduce to only `flower` and `wilting` being in sprximity.
             "text": "A flower wilting under the sun, unlike a sunflower"
@@ -299,7 +299,7 @@ fn test_proximity_split_word() {
     let SearchResult { documents_ids, .. } = s.execute().unwrap();
     insta::assert_snapshot!(format!("{documents_ids:?}"), @"[2, 4, 5, 1, 3]");
     let texts = collect_field_values(&index, &txn, "text", &documents_ids);
-    // TODO: "2" and "4" should be swapped ideally
+    // "2" and "4" should be swapped ideally
     insta::assert_debug_snapshot!(texts, @r###"
     [
         "\"Sun Flower sounds like the title of a painting, maybe about a flower wilting under the heat.\"",
@@ -316,7 +316,7 @@ fn test_proximity_split_word() {
     let SearchResult { documents_ids, .. } = s.execute().unwrap();
     insta::assert_snapshot!(format!("{documents_ids:?}"), @"[2, 4, 1]");
     let texts = collect_field_values(&index, &txn, "text", &documents_ids);
-    // TODO: "2" and "4" should be swapped ideally
+    // "2" and "4" should be swapped ideally
     insta::assert_debug_snapshot!(texts, @r###"
     [
         "\"Sun Flower sounds like the title of a painting, maybe about a flower wilting under the heat.\"",
@@ -341,7 +341,7 @@ fn test_proximity_split_word() {
     let SearchResult { documents_ids, .. } = s.execute().unwrap();
     insta::assert_snapshot!(format!("{documents_ids:?}"), @"[2, 4, 1]");
     let texts = collect_field_values(&index, &txn, "text", &documents_ids);
-    // TODO: "2" and "4" should be swapped ideally
+    // "2" and "4" should be swapped ideally
     insta::assert_debug_snapshot!(texts, @r###"
     [
         "\"Sun Flower sounds like the title of a painting, maybe about a flower wilting under the heat.\"",

milli/src/search/new/tests/proximity_typo.rs

@@ -2,9 +2,8 @@
 This module tests the interactions between the proximity and typo ranking rules.
 
 The proximity ranking rule should transform the query graph such that it
-only contains the word pairs that it used to compute its bucket.
-
-TODO: This is not currently implemented.
+only contains the word pairs that it used to compute its bucket, but this is not currently
+implemented.
 */
 
 use crate::index::tests::TempIndex;
@@ -64,7 +63,7 @@ fn test_trap_basic() {
     let SearchResult { documents_ids, .. } = s.execute().unwrap();
     insta::assert_snapshot!(format!("{documents_ids:?}"), @"[0, 1]");
     let texts = collect_field_values(&index, &txn, "text", &documents_ids);
-    // TODO: this is incorrect, 1 should come before 0
+    // This is incorrect, 1 should come before 0
     insta::assert_debug_snapshot!(texts, @r###"
     [
         "\"summer. holiday. sommer holidty\"",

milli/src/search/new/tests/typo.rs

@@ -571,8 +571,8 @@ fn test_typo_synonyms() {
     s.terms_matching_strategy(TermsMatchingStrategy::All);
     s.query("the fast brownish fox jumps over the lackadaisical dog");
 
-    // TODO: is this correct? interaction of ngrams + synonyms means that the
-    // multi-word synonyms end up having a typo cost. This is probably not what we want.
+    // The interaction of ngrams + synonyms means that the multi-word synonyms end up having a typo cost.
+    // This is probably not what we want.
     let SearchResult { documents_ids, .. } = s.execute().unwrap();
     insta::assert_snapshot!(format!("{documents_ids:?}"), @"[21, 0, 22]");
     let texts = collect_field_values(&index, &txn, "text", &documents_ids);

milli/src/snapshot_tests.rs

@@ -318,7 +318,7 @@ pub fn snap_field_distributions(index: &Index) -> String {
     let rtxn = index.read_txn().unwrap();
     let mut snap = String::new();
     for (field, count) in index.field_distribution(&rtxn).unwrap() {
-        writeln!(&mut snap, "{field:<16} {count:<6}").unwrap();
+        writeln!(&mut snap, "{field:<16} {count:<6} |").unwrap();
     }
     snap
 }
@@ -328,7 +328,7 @@ pub fn snap_fields_ids_map(index: &Index) -> String {
     let mut snap = String::new();
     for field_id in fields_ids_map.ids() {
         let name = fields_ids_map.name(field_id).unwrap();
-        writeln!(&mut snap, "{field_id:<3} {name:<16}").unwrap();
+        writeln!(&mut snap, "{field_id:<3} {name:<16} |").unwrap();
     }
     snap
 }

milli/src/snapshots/index.rs/initial_field_distribution/1/field_distribution.snap

@@ -1,7 +1,7 @@
 ---
 source: milli/src/index.rs
 ---
-age              1     
-id               2     
-name             2     
+age              1      |
+id               2      |
+name             2      |
 

milli/src/snapshots/index.rs/initial_field_distribution/field_distribution.snap

@@ -1,7 +1,7 @@
 ---
 source: milli/src/index.rs
 ---
-age              1     
-id               2     
-name             2     
+age              1      |
+id               2      |
+name             2      |
 

milli/src/update/delete_documents.rs

@@ -71,7 +71,6 @@ impl std::fmt::Display for DeletionStrategy {
 pub(crate) struct DetailedDocumentDeletionResult {
     pub deleted_documents: u64,
     pub remaining_documents: u64,
-    pub soft_deletion_used: bool,
 }
 
 impl<'t, 'u, 'i> DeleteDocuments<'t, 'u, 'i> {
@@ -108,11 +107,8 @@ impl<'t, 'u, 'i> DeleteDocuments<'t, 'u, 'i> {
         Some(docid)
     }
     pub fn execute(self) -> Result<DocumentDeletionResult> {
-        let DetailedDocumentDeletionResult {
-            deleted_documents,
-            remaining_documents,
-            soft_deletion_used: _,
-        } = self.execute_inner()?;
+        let DetailedDocumentDeletionResult { deleted_documents, remaining_documents } =
+            self.execute_inner()?;
 
         Ok(DocumentDeletionResult { deleted_documents, remaining_documents })
     }
@@ -133,7 +129,6 @@ impl<'t, 'u, 'i> DeleteDocuments<'t, 'u, 'i> {
             return Ok(DetailedDocumentDeletionResult {
                 deleted_documents: 0,
                 remaining_documents: 0,
-                soft_deletion_used: false,
             });
         }
 
@@ -149,7 +144,6 @@ impl<'t, 'u, 'i> DeleteDocuments<'t, 'u, 'i> {
             return Ok(DetailedDocumentDeletionResult {
                 deleted_documents: current_documents_ids_len,
                 remaining_documents,
-                soft_deletion_used: false,
             });
         }
 
@@ -218,7 +212,6 @@ impl<'t, 'u, 'i> DeleteDocuments<'t, 'u, 'i> {
             return Ok(DetailedDocumentDeletionResult {
                 deleted_documents: self.to_delete_docids.len(),
                 remaining_documents: documents_ids.len(),
-                soft_deletion_used: true,
             });
         }
 
@@ -441,7 +434,6 @@ impl<'t, 'u, 'i> DeleteDocuments<'t, 'u, 'i> {
         Ok(DetailedDocumentDeletionResult {
             deleted_documents: self.to_delete_docids.len(),
             remaining_documents: documents_ids.len(),
-            soft_deletion_used: false,
         })
     }
 

milli/src/update/index_documents/helpers/clonable_mmap.rs

@@ -2,7 +2,7 @@ use std::sync::Arc;
 
 use memmap2::Mmap;
 
-/// Wrapper around Mmap allowing to virtualy clone grenad-chunks
+/// Wrapper around Mmap allowing to virtually clone grenad-chunks
 /// in a parallel process like the indexing.
 #[derive(Debug, Clone)]
 pub struct ClonableMmap {

milli/src/update/index_documents/mod.rs

@@ -236,7 +236,7 @@ where
             primary_key,
             fields_ids_map,
             field_distribution,
-            mut external_documents_ids,
+            new_external_documents_ids,
             new_documents_ids,
             replaced_documents_ids,
             documents_count,
@@ -363,9 +363,6 @@ where
             deletion_builder.delete_documents(&replaced_documents_ids);
             let deleted_documents_result = deletion_builder.execute_inner()?;
             debug!("{} documents actually deleted", deleted_documents_result.deleted_documents);
-            if !deleted_documents_result.soft_deletion_used {
-                external_documents_ids.delete_soft_deleted_documents_ids_from_fsts()?;
-            }
         }
 
         let index_documents_ids = self.index.documents_ids(self.wtxn)?;
@@ -445,6 +442,9 @@ where
         self.index.put_primary_key(self.wtxn, &primary_key)?;
 
         // We write the external documents ids into the main database.
+        let mut external_documents_ids = self.index.external_documents_ids(self.wtxn)?;
+        external_documents_ids.insert_ids(&new_external_documents_ids)?;
+        let external_documents_ids = external_documents_ids.into_static();
         self.index.put_external_documents_ids(self.wtxn, &external_documents_ids)?;
 
         let all_documents_ids = index_documents_ids | new_documents_ids;
@@ -2514,4 +2514,170 @@ mod tests {
         db_snap!(index, word_fid_docids, 3, @"4c2e2a1832e5802796edc1638136d933");
         db_snap!(index, word_position_docids, 3, @"74f556b91d161d997a89468b4da1cb8f");
     }
+
+    #[test]
+    fn reproduce_the_bug() {
+        /*
+            [milli/examples/fuzz.rs:69] &batches = [
+            Batch(
+                [
+                    AddDoc(
+                        { "id": 1, "doggo": "bernese" }, => internal 0
+                    ),
+                ],
+            ),
+            Batch(
+                [
+                    DeleteDoc(
+                        1, => delete internal 0
+                    ),
+                    AddDoc(
+                        { "id": 0, "catto": "jorts" }, => internal 1
+                    ),
+                ],
+            ),
+            Batch(
+                [
+                    AddDoc(
+                        { "id": 1, "catto": "jorts" }, => internal 2
+                    ),
+                ],
+            ),
+        ]
+        */
+        let mut index = TempIndex::new();
+        index.index_documents_config.deletion_strategy = DeletionStrategy::AlwaysHard;
+
+        // START OF BATCH
+
+        println!("--- ENTERING BATCH 1");
+
+        let mut wtxn = index.write_txn().unwrap();
+
+        let builder = IndexDocuments::new(
+            &mut wtxn,
+            &index,
+            &index.indexer_config,
+            index.index_documents_config.clone(),
+            |_| (),
+            || false,
+        )
+        .unwrap();
+
+        // OP
+
+        let documents = documents!([
+            { "id": 1, "doggo": "bernese" },
+        ]);
+        let (builder, added) = builder.add_documents(documents).unwrap();
+        insta::assert_display_snapshot!(added.unwrap(), @"1");
+
+        // FINISHING
+        let addition = builder.execute().unwrap();
+        insta::assert_debug_snapshot!(addition, @r###"
+        DocumentAdditionResult {
+            indexed_documents: 1,
+            number_of_documents: 1,
+        }
+        "###);
+        wtxn.commit().unwrap();
+
+        db_snap!(index, documents, @r###"
+        {"id":1,"doggo":"bernese"}
+        "###);
+        db_snap!(index, external_documents_ids, @r###"
+        soft:
+        hard:
+        1                        0
+        "###);
+
+        // A first batch of documents has been inserted
+
+        // BATCH 2
+
+        println!("--- ENTERING BATCH 2");
+
+        let mut wtxn = index.write_txn().unwrap();
+
+        let builder = IndexDocuments::new(
+            &mut wtxn,
+            &index,
+            &index.indexer_config,
+            index.index_documents_config.clone(),
+            |_| (),
+            || false,
+        )
+        .unwrap();
+
+        let (builder, removed) = builder.remove_documents(vec![S("1")]).unwrap();
+        insta::assert_display_snapshot!(removed.unwrap(), @"1");
+
+        let documents = documents!([
+            { "id": 0, "catto": "jorts" },
+        ]);
+        let (builder, added) = builder.add_documents(documents).unwrap();
+        insta::assert_display_snapshot!(added.unwrap(), @"1");
+
+        let addition = builder.execute().unwrap();
+        insta::assert_debug_snapshot!(addition, @r###"
+        DocumentAdditionResult {
+            indexed_documents: 1,
+            number_of_documents: 1,
+        }
+        "###);
+        wtxn.commit().unwrap();
+
+        db_snap!(index, documents, @r###"
+        {"id":0,"catto":"jorts"}
+        "###);
+
+        db_snap!(index, external_documents_ids, @r###"
+        soft:
+        hard:
+        0                        1
+        "###);
+
+        db_snap!(index, soft_deleted_documents_ids, @"[]");
+
+        // BATCH 3
+
+        println!("--- ENTERING BATCH 3");
+
+        let mut wtxn = index.write_txn().unwrap();
+
+        let builder = IndexDocuments::new(
+            &mut wtxn,
+            &index,
+            &index.indexer_config,
+            index.index_documents_config.clone(),
+            |_| (),
+            || false,
+        )
+        .unwrap();
+
+        let documents = documents!([
+            { "id": 1, "catto": "jorts" },
+        ]);
+        let (builder, added) = builder.add_documents(documents).unwrap();
+        insta::assert_display_snapshot!(added.unwrap(), @"1");
+
+        let addition = builder.execute().unwrap();
+        insta::assert_debug_snapshot!(addition, @r###"
+        DocumentAdditionResult {
+            indexed_documents: 1,
+            number_of_documents: 2,
+        }
+        "###);
+        wtxn.commit().unwrap();
+
+        db_snap!(index, documents, @r###"
+        {"id":1,"catto":"jorts"}
+        {"id":0,"catto":"jorts"}
+        "###);
+
+        // Ensuring all the returned IDs actually exists
+        let rtxn = index.read_txn().unwrap();
+        let res = index.search(&rtxn).execute().unwrap();
+        index.documents(&rtxn, res.documents_ids).unwrap();
+    }
 }

milli/src/update/index_documents/transform.rs

@@ -21,15 +21,14 @@ use crate::error::{Error, InternalError, UserError};
 use crate::index::{db_name, main_key};
 use crate::update::{AvailableDocumentsIds, ClearDocuments, UpdateIndexingStep};
 use crate::{
-    ExternalDocumentsIds, FieldDistribution, FieldId, FieldIdMapMissingEntry, FieldsIdsMap, Index,
-    Result, BEU32,
+    FieldDistribution, FieldId, FieldIdMapMissingEntry, FieldsIdsMap, Index, Result, BEU32,
 };
 
 pub struct TransformOutput {
     pub primary_key: String,
     pub fields_ids_map: FieldsIdsMap,
     pub field_distribution: FieldDistribution,
-    pub external_documents_ids: ExternalDocumentsIds<'static>,
+    pub new_external_documents_ids: fst::Map<Cow<'static, [u8]>>,
     pub new_documents_ids: RoaringBitmap,
     pub replaced_documents_ids: RoaringBitmap,
     pub documents_count: usize,
@@ -568,8 +567,6 @@ impl<'a, 'i> Transform<'a, 'i> {
             }))?
             .to_string();
 
-        let mut external_documents_ids = self.index.external_documents_ids(wtxn)?;
-
         // We create a final writer to write the new documents in order from the sorter.
         let mut writer = create_writer(
             self.indexer_settings.chunk_compression_type,
@@ -651,13 +648,12 @@ impl<'a, 'i> Transform<'a, 'i> {
             fst_new_external_documents_ids_builder.insert(key, value)
         })?;
         let new_external_documents_ids = fst_new_external_documents_ids_builder.into_map();
-        external_documents_ids.insert_ids(&new_external_documents_ids)?;
 
         Ok(TransformOutput {
             primary_key,
             fields_ids_map: self.fields_ids_map,
             field_distribution,
-            external_documents_ids: external_documents_ids.into_static(),
+            new_external_documents_ids: new_external_documents_ids.map_data(Cow::Owned).unwrap(),
             new_documents_ids: self.new_documents_ids,
             replaced_documents_ids: self.replaced_documents_ids,
             documents_count: self.documents_count,
@@ -691,7 +687,8 @@ impl<'a, 'i> Transform<'a, 'i> {
         let new_external_documents_ids = {
             let mut external_documents_ids = self.index.external_documents_ids(wtxn)?;
             external_documents_ids.delete_soft_deleted_documents_ids_from_fsts()?;
-            external_documents_ids
+            // This call should be free and can't fail since the previous method merged both fsts.
+            external_documents_ids.into_static().to_fst()?.into_owned()
         };
 
         let documents_ids = self.index.documents_ids(wtxn)?;
@@ -776,7 +773,7 @@ impl<'a, 'i> Transform<'a, 'i> {
             primary_key,
             fields_ids_map: new_fields_ids_map,
             field_distribution,
-            external_documents_ids: new_external_documents_ids.into_static(),
+            new_external_documents_ids,
             new_documents_ids: documents_ids,
             replaced_documents_ids: RoaringBitmap::default(),
             documents_count,