Merge pull request #342 from meilisearch/update-lock

Update the lock file

.dockerignore

@@ -0,0 +1,5 @@
+target
+Dockerfile
+.dockerignore
+.git
+.gitignore

.github/workflows/README.md

@@ -0,0 +1,11 @@
+# GitHub actions workflow for MeiliDB
+
+> **Note:**
+
+> - We do not use [cache](https://github.com/actions/cache) yet but we could use it to speed up CI
+
+## Workflow
+
+- On each pull request, we are triggering `cargo test`.
+- On each commit on master, we are building the latest docker image.
+- On each tag, we are building the tagged docker image and the binaries for MacOS & Ubuntu.

.github/workflows/publish-binaries.yml

@@ -0,0 +1,36 @@
+name: Publish binaries to GitHub release
+
+on:
+  push:
+    tags:
+      - '*'
+
+jobs:
+  publish:
+    name: Publish for ${{ matrix.os }}
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        os: [ubuntu-latest, macos-latest]
+        include:
+          - os: ubuntu-latest
+            artifact_name: meilisearch
+            asset_name: meilisearch-linux-amd64
+          - os: macos-latest
+            artifact_name: meilisearch
+            asset_name: meilisearch-macos-amd64
+
+    steps:
+    - uses: hecrj/setup-rust-action@master
+      with:
+        rust-version: stable
+    - uses: actions/checkout@v1
+    - name: Build
+      run: cargo build --release --locked
+    - name: Upload binaries to release
+      uses: svenstaro/upload-release-action@v1-release
+      with:
+        repo_token: ${{ secrets.GITHUB_TOKEN }}
+        file: target/release/${{ matrix.artifact_name }}
+        asset_name: ${{ matrix.asset_name }}
+        tag: ${{ github.ref }}

.github/workflows/publish-docker-latest.yml

@@ -0,0 +1,19 @@
+---
+on:
+  push:
+    branches:
+      - master
+
+name: Publish latest image to Docker Hub
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v1
+      - name: Publish to Registry
+        uses: elgohr/Publish-Docker-Github-Action@master
+        with:
+          name: getmeili/meilisearch
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_PASSWORD }}

.github/workflows/publish-docker-tag.yml

@@ -0,0 +1,20 @@
+---
+on:
+  push:
+    tags:
+      - '*'
+
+name: Publish tagged image to Docker Hub
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v1
+      - name: Publish to Registry
+        uses: elgohr/Publish-Docker-Github-Action@master
+        with:
+          name: getmeili/meilisearch
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_PASSWORD }}
+          tag_names: true

.github/workflows/test.yml

@@ -0,0 +1,21 @@
+---
+on: [pull_request]
+
+name: Cargo test
+
+jobs:
+  check:
+    name: MeiliSearch
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v1
+      - uses: actions-rs/toolchain@v1
+        with:
+          profile: minimal
+          toolchain: stable
+          override: true
+      - name: Run cargo test
+        uses: actions-rs/cargo@v1
+        with:
+          command: test
+          args: --locked

.gitignore

@@ -1,5 +1,4 @@
 /target
-Cargo.lock
 **/*.csv
 **/*.json_lines
 **/*.rs.bk

Cargo.toml

@@ -1,8 +1,10 @@
 [workspace]
 members = [
-    "meilidb-core",
-    "meilidb-schema",
-    "meilidb-tokenizer",
+    "meilisearch-core",
+    "meilisearch-http",
+    "meilisearch-schema",
+    "meilisearch-tokenizer",
+    "meilisearch-types",
 ]
 
 [profile.release]

Dockerfile

@@ -0,0 +1,29 @@
+# Compile
+FROM    alpine:3.10 AS compiler
+
+RUN     apk update --quiet
+RUN     apk add curl
+RUN     apk add build-base
+RUN     apk add libressl-dev
+
+RUN     curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y
+
+WORKDIR /meilisearch
+
+COPY    . .
+
+ENV     RUSTFLAGS="-C target-feature=-crt-static"
+
+RUN     $HOME/.cargo/bin/cargo build --release
+
+# Run
+FROM    alpine:3.10
+
+RUN     apk update --quiet
+RUN     apk add libressl
+RUN     apk add build-base
+
+COPY    --from=compiler /meilisearch/target/release/meilisearch .
+
+ENV     MEILI_HTTP_ADDR 0.0.0.0:8080
+CMD     ./meilisearch

LICENSE

@@ -1,13 +1,21 @@
-“Commons Clause” License Condition v1.0
+MIT License
 
-The Software is provided to you by the Licensor under the License, as defined below, subject to the following condition.
+Copyright (c) [year] [fullname]
 
-Without limiting other conditions in the License, the grant of rights under the License will not include, and the License does not grant to you, the right to Sell the Software.
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
 
-For purposes of the foregoing, “Sell” means practicing any or all of the rights granted to you under the License to provide to third parties, for a fee or other consideration (including without limitation fees for hosting or consulting/ support services related to the Software), a product or service whose value derives, entirely or substantially, from the functionality of the Software. Any license notice or attribution required by the License must also include this Commons Clause License Condition notice.
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
 
-Software: MeiliDB
-
-License: MIT
-
-Licensor: MEILI SAS
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

README.md

@@ -1,41 +1,117 @@
-# MeiliDB
+# MeiliSearch
 
-[![Build Status](https://dev.azure.com/thomas0884/thomas/_apis/build/status/meilisearch.MeiliDB?branchName=master)](https://dev.azure.com/thomas0884/thomas/_build/latest?definitionId=1&branchName=master)
-[![dependency status](https://deps.rs/repo/github/meilisearch/MeiliDB/status.svg)](https://deps.rs/repo/github/meilisearch/MeiliDB)
+[![Build Status](https://github.com/meilisearch/MeiliSearch/workflows/Cargo%20test/badge.svg)](https://github.com/meilisearch/MeiliSearch/actions)
+[![dependency status](https://deps.rs/repo/github/meilisearch/MeiliSearch/status.svg)](https://deps.rs/repo/github/meilisearch/MeiliSearch)
 [![License](https://img.shields.io/badge/license-commons%20clause-lightgrey)](https://commonsclause.com/)
 
-A _full-text search database_ based on the fast [LMDB key-value store](https://en.wikipedia.org/wiki/Lightning_Memory-Mapped_Database).
+Ultra relevant and instant full-text search API.
+
+MeiliSearch is a powerful, fast, open-source, easy to use and deploy search engine. The search and indexation are fully customizable and handles features like typo-tolerance, filters, and ranking.
 
 ## Features
 
-- Provides [6 default ranking criteria](https://github.com/meilisearch/MeiliDB/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilidb-core/src/criterion/mod.rs#L107-L113) used to [bucket sort](https://en.wikipedia.org/wiki/Bucket_sort) documents
-- Accepts [custom criteria](https://github.com/meilisearch/MeiliDB/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilidb-core/src/criterion/mod.rs#L24-L33) and can apply them in any custom order
-- Support [ranged queries](https://github.com/meilisearch/MeiliDB/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilidb-core/src/query_builder.rs#L283), useful for paginating results
-- Can [distinct](https://github.com/meilisearch/MeiliDB/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilidb-core/src/query_builder.rs#L265-L270) and [filter](https://github.com/meilisearch/MeiliDB/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilidb-core/src/query_builder.rs#L246-L259) returned documents based on context defined rules
-- Searches for [concatenated](https://github.com/meilisearch/MeiliDB/pull/164) and [splitted query words](https://github.com/meilisearch/MeiliDB/pull/232) to improve the search quality.
-- Can store complete documents or only [user schema specified fields](https://github.com/meilisearch/MeiliDB/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilidb-schema/src/lib.rs#L265-L279)
-- The [default tokenizer](https://github.com/meilisearch/MeiliDB/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilidb-tokenizer/src/lib.rs) can index latin and kanji based languages
-- Returns [the matching text areas](https://github.com/meilisearch/MeiliDB/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilidb-core/src/lib.rs#L66-L88), useful to highlight matched words in results
-- Accepts query time search config like the [searchable attributes](https://github.com/meilisearch/MeiliDB/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilidb-core/src/query_builder.rs#L272-L275)
-- Supports [runtime incremental indexing](https://github.com/meilisearch/MeiliDB/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilidb-core/src/store/mod.rs#L143-L173)
+- Provides [6 default ranking criteria](https://github.com/meilisearch/MeiliSearch/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilisearch-core/src/criterion/mod.rs#L107-L113) used to [bucket sort](https://en.wikipedia.org/wiki/Bucket_sort) documents
+- Accepts [custom criteria](https://github.com/meilisearch/MeiliSearch/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilisearch-core/src/criterion/mod.rs#L24-L33) and can apply them in any custom order
+- Support [ranged queries](https://github.com/meilisearch/MeiliSearch/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilisearch-core/src/query_builder.rs#L283), useful for paginating results
+- Can [distinct](https://github.com/meilisearch/MeiliSearch/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilisearch-core/src/query_builder.rs#L265-L270) and [filter](https://github.com/meilisearch/MeiliSearch/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilisearch-core/src/query_builder.rs#L246-L259) returned documents based on context defined rules
+- Searches for [concatenated](https://github.com/meilisearch/MeiliSearch/pull/164) and [splitted query words](https://github.com/meilisearch/MeiliSearch/pull/232) to improve the search quality.
+- Can store complete documents or only [user schema specified fields](https://github.com/meilisearch/MeiliSearch/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilisearch-schema/src/lib.rs#L265-L279)
+- The [default tokenizer](https://github.com/meilisearch/MeiliSearch/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilisearch-tokenizer/src/lib.rs) can index latin and kanji based languages
+- Returns [the matching text areas](https://github.com/meilisearch/MeiliSearch/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilisearch-core/src/lib.rs#L66-L88), useful to highlight matched words in results
+- Accepts query time search config like the [searchable attributes](https://github.com/meilisearch/MeiliSearch/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilisearch-core/src/query_builder.rs#L272-L275)
+- Supports [runtime incremental indexing](https://github.com/meilisearch/MeiliSearch/blob/dc5c42821e1340e96cb90a3da472264624a26326/meilisearch-core/src/store/mod.rs#L143-L173)
 
 
 
-It uses [LMDB](https://en.wikipedia.org/wiki/Lightning_Memory-Mapped_Database) as the internal key-value store. The key-value store allows us to handle updates and queries with small memory and CPU overheads. The whole ranking system is [data oriented](https://github.com/meilisearch/MeiliDB/issues/82) and provides great performances.
+It uses [LMDB](https://en.wikipedia.org/wiki/Lightning_Memory-Mapped_Database) as the internal key-value store. The key-value store allows us to handle updates and queries with small memory and CPU overheads. The whole ranking system is [data oriented](https://github.com/meilisearch/MeiliSearch/issues/82) and provides great performances.
 
 You can [read the deep dive](deep-dive.md) if you want more information on the engine, it describes the whole process of generating updates and handling queries or you can take a look at the [typos and ranking rules](typos-ranking-rules.md) if you want to know the default rules used to sort the documents.
 
-We will be proud if you submit issues and pull requests. You can help to grow this project and start contributing by checking [issues tagged "good-first-issue"](https://github.com/meilisearch/MeiliDB/issues?q=is%3Aissue+is%3Aopen+label%3A%22good+first+issue%22). It is a good start!
+We will be glad if you submit issues and pull requests. You can help to grow this project and start contributing by checking [issues tagged "good-first-issue"](https://github.com/meilisearch/MeiliSearch/issues?q=is%3Aissue+is%3Aopen+label%3A%22good+first+issue%22). It is a good start!
 
-The project is only a library yet. It means that there is no binary provided yet. To get started, you can check the examples wich are made to work with the data located in the `datasets/` folder.
+[![crates.io demo gif](misc/crates-io-demo.gif)](https://crates.meilisearch.com)
 
-MeiliDB will be a binary in a near future so you will be able to use it as a database out-of-the-box. We should be able to query it using HTTP. This is our current goal, [see the milestones](https://github.com/meilisearch/MeiliDB/milestones). In the end, the binary will be a bunch of network protocols and wrappers around the library - which will also be published on [crates.io](https://crates.io). Both the binary and the library will follow the same update cycle.
+> Meili helps the Rust community find crates on [crates.meilisearch.com](https://crates.meilisearch.com)
+
+
+
+## Quick Start
+
+You can deploy your own instant, relevant and typo-tolerant MeiliSearch search engine by yourself too.
+Something similar to the demo above can be achieve by following these little three steps first.
+You will need to create your own web front display to make it pretty though.
+
+### Deploy the Server
+
+If you have not installed Rust and its package manager `cargo` yet, go to [the installation page](https://www.rust-lang.org/tools/install).<br/>
+You can deploy the server on your own machine, it will listen to HTTP requests on the 8080 port by default.
+
+```bash
+cargo run --release
+```
+
+For more logs during the execution, run:
+```bash
+RUST_LOG=info cargo run --release
+```
+
+### Create an Index and Upload Some Documents
+
+MeiliSearch can serve multiple indexes, with different kinds of documents,
+therefore, it is required to create the index before sending documents to it.
+
+```bash
+curl -i -X POST 'http://127.0.0.1:8080/indexes' --data '{ "name": "Movies", "uid": "movies" }'
+```
+
+Now that the server knows about our brand new index, we can send it data.
+We provided you a little dataset, it is available in the `datasets/` directory.
+
+```bash
+curl -i -X POST 'http://127.0.0.1:8080/indexes/movies/documents' \
+  --header 'content-type: application/json' \
+  --data @datasets/movies/movies.json
+```
+
+### Search for Documents
+
+The search engine is now aware of our documents and can serve those via our HTTP server again.
+The [`jq` command line tool](https://stedolan.github.io/jq/) can greatly help you read the server responses.
+
+```bash
+curl 'http://127.0.0.1:8080/indexes/movies/search?q=botman'
+```
+
+```json
+{
+  "hits": [
+    {
+      "id": "29751",
+      "title": "Batman Unmasked: The Psychology of the Dark Knight",
+      "poster": "https://image.tmdb.org/t/p/w1280/jjHu128XLARc2k4cJrblAvZe0HE.jpg",
+      "overview": "Delve into the world of Batman and the vigilante justice tha",
+      "release_date": "2008-07-15"
+    },
+    {
+      "id": "471474",
+      "title": "Batman: Gotham by Gaslight",
+      "poster": "https://image.tmdb.org/t/p/w1280/7souLi5zqQCnpZVghaXv0Wowi0y.jpg",
+      "overview": "ve Victorian Age Gotham City, Batman begins his war on crime",
+      "release_date": "2018-01-12"
+    }
+  ],
+  "offset": 0,
+  "limit": 2,
+  "processingTimeMs": 1,
+  "query": "botman"
+}
+```
 
 
 
 ## Performances
 
-With a database composed of _100 353_ documents with _352_ attributes each and _3_ of them indexed.
+With a dataset composed of _100 353_ documents with _352_ attributes each and _3_ of them indexed.
 So more than _300 000_ fields indexed for _35 million_ stored we can handle more than _2.8k req/sec_ with an average response time of _9 ms_ on an Intel i7-7700 (8) @ 4.2GHz.
 
 Requests are made using [wrk](https://github.com/wg/wrk) and scripted to simulate real users queries.
@@ -51,20 +127,24 @@ Requests/sec:   2806.46
 Transfer/sec:    759.17KB
 ```
 
+We also indexed a dataset containing something like _12 millions_ cities names in _24 minutes_ on a machine with _8 cores_, _64 GB of RAM_ and a _300 GB NMVe_ SSD.<br/>
+The resulting database was _16 GB_ and search results were between _30 ms_ and _4 seconds_ for short prefix queries.
+
 ### Notes
 
 With Rust 1.32 the allocator has been [changed to use the system allocator](https://blog.rust-lang.org/2019/01/17/Rust-1.32.0.html#jemalloc-is-removed-by-default).
 We have seen much better performances when [using jemalloc as the global allocator](https://github.com/alexcrichton/jemallocator#documentation).
 
-## Usage and examples
+## Usage and Examples
 
-Currently MeiliDB do not provide an http server but you can run the example binary.
+MeiliSearch also provides an example binary that is mostly used for features testing.
+Notice that the example binary is faster to index data as it does read direct CSV files and not JSON HTTP payloads.
 
-The _index_ subcommand has been made to create an index and inject documents into it. Using the command line below, the index will be named _movies_ and the _19 700_ movies of the `datasets/` will be injected in MeiliDB.
+The _index_ subcommand has been made to create an index and inject documents into it. Using the command line below, the index will be named _movies_ and the _19 700_ movies of the `datasets/` will be injected in MeiliSearch.
 
 ```bash
 cargo run --release --example from_file -- \
-    index example.mdb datasets/movies/data.csv \
+    index example.mdb datasets/movies/movies.csv \
     --schema datasets/movies/schema.toml
 ```
 
@@ -72,8 +152,14 @@ Once the first command is done, you can query the freshly created _movies_ index
 
 ```bash
 cargo run --release --example from_file -- \
-    search example.mdb
-    --number 4 \
+    search example.mdb \
+    --number-results 4 \
     --filter '!adult' \
     id popularity adult original_title
 ```
+
+### Analytic Events
+
+We send events to our Amplitude instance to be aware of the number of people who use MeiliSearch.<br/>
+We only send the platform on which the server runs once by day. No other information is sent.<br/>
+If you do not want us to send events, you can disable these analytics by using the `MEILI_NO_ANALYTICS` env variable.

azure-pipelines.yml

@@ -1,52 +0,0 @@
----
-trigger:
-  branches:
-    include: [ master ]
-
-pr: [ master ]
-
-jobs:
-  - job: test
-    pool:
-      vmImage: 'Ubuntu 16.04'
-    container: tpayet/chiquitita:latest
-    steps:
-      - script: |
-          curl https://sh.rustup.rs -sSf | sh -s -- -y --default-toolchain nightly
-          $HOME/.cargo/bin/rustup component add rustfmt
-        displayName: 'Install rustc and components'
-      - script: |
-          $HOME/.cargo/bin/cargo check
-        displayName: 'Check MeiliDB'
-      - script: |
-          $HOME/.cargo/bin/cargo test
-        displayName: 'Test MeiliDB'
-      - script: |
-          $HOME/.cargo/bin/cargo fmt --all -- --check
-        displayName: 'Fmt MeiliDB'
-
-  - job: build
-    dependsOn:
-      - test
-    condition: succeeded()
-    pool:
-      vmImage: 'Ubuntu 16.04'
-    container: tpayet/chiquitita:latest
-    steps:
-      - script: |
-          curl https://sh.rustup.rs -sSf | sh -s -- -y --default-toolchain nightly
-          $HOME/.cargo/bin/rustup component add rustfmt
-        displayName: 'Install rustc and components'
-      - script: |
-          $HOME/.cargo/bin/cargo build --release
-        displayName: 'Build MeiliDB'
-      - task: CopyFiles@2
-        inputs:
-          contents: '$(System.DefaultWorkingDirectory)/target/release/libmeilidb.rlib'
-          targetFolder: $(Build.ArtifactStagingDirectory)
-        displayName: 'Copy build'
-      - task: PublishBuildArtifacts@1
-        inputs:
-          artifactName: libmeilidb.rlib
-        displayName: 'Upload artifacts'
-

datasets/movies/README.md

@@ -1 +1 @@
-_datas in movies.csv are from https://www.themoviedb.org/_
+_datas in movies.csv are from https://www.themoviedb.org/_

deep-dive.md

@@ -1,8 +1,8 @@
-# A deep dive in MeiliDB
+# A deep dive in MeiliSearch
 
 On the 15 of May 2019.
 
-MeiliDB is a full text search engine based on a final state transducer named [fst](https://github.com/BurntSushi/fst) and a key-value store named [sled](https://github.com/spacejam/sled). The goal of a search engine is to store data and to respond to queries as accurate and fast as possible. To achieve this it must save the matching words in an [inverted index](https://en.wikipedia.org/wiki/Inverted_index).
+MeiliSearch is a full text search engine based on a final state transducer named [fst](https://github.com/BurntSushi/fst) and a key-value store named [sled](https://github.com/spacejam/sled). The goal of a search engine is to store data and to respond to queries as accurate and fast as possible. To achieve this it must save the matching words in an [inverted index](https://en.wikipedia.org/wiki/Inverted_index).
 
 <!-- MarkdownTOC autolink="true" -->
 
@@ -22,7 +22,7 @@ MeiliDB is a full text search engine based on a final state transducer named [fs
 
 ## Where is the data stored?
 
-MeiliDB is entirely backed by a key-value store like any good database (i.e. Postgres, MySQL). This brings a great flexibility in the way documents can be stored and updates handled along time.
+MeiliSearch is entirely backed by a key-value store like any good database (i.e. Postgres, MySQL). This brings a great flexibility in the way documents can be stored and updates handled along time.
 
 [sled will brings some](https://github.com/spacejam/sled/tree/434533332a3f485e6d2e467023be0a0b55d3a1af#plans) of the [A.C.I.D. properties](https://en.wikipedia.org/wiki/ACID_(computer_science)) to help us be sure the saved data is consistent.
 
@@ -34,7 +34,7 @@ It contain the inverted word index, the schema and the documents fields.
 
 ### The inverted word index
 
-[The inverted word index](https://github.com/meilisearch/MeiliDB/blob/3db823de002243004612e36a19b4578d800dab97/meilidb-data/src/database/words_index.rs) is a sled Tree dedicated to store and give access to all documents that contains a specific word. The information stored under the word is simply a big ordered array of where in the document the word has been found. In other word, a big list of [`DocIndex`](https://github.com/meilisearch/MeiliDB/blob/3db823de002243004612e36a19b4578d800dab97/meilidb-core/src/lib.rs#L35-L51).
+[The inverted word index](https://github.com/meilisearch/MeiliSearch/blob/3db823de002243004612e36a19b4578d800dab97/meilisearch-data/src/database/words_index.rs) is a sled Tree dedicated to store and give access to all documents that contains a specific word. The information stored under the word is simply a big ordered array of where in the document the word has been found. In other word, a big list of [`DocIndex`](https://github.com/meilisearch/MeiliSearch/blob/3db823de002243004612e36a19b4578d800dab97/meilisearch-core/src/lib.rs#L35-L51).
 
 #### A final state transducer
 
@@ -42,27 +42,27 @@ _...also abbreviated fst_
 
 This is the first entry point of the engine, you can read more about how it work with the beautiful blog post of @BurntSushi, [Index 1,600,000,000 Keys with Automata and Rust](https://blog.burntsushi.net/transducers/).
 
-To make it short it is a powerful way to store all the words that are present in the indexed documents. You construct it by giving it all the words you want to index. When you want to search in it you can provide any automaton you want, in MeiliDB [a custom levenshtein automaton](https://github.com/tantivy-search/levenshtein-automata/) is used.
+To make it short it is a powerful way to store all the words that are present in the indexed documents. You construct it by giving it all the words you want to index. When you want to search in it you can provide any automaton you want, in MeiliSearch [a custom levenshtein automaton](https://github.com/tantivy-search/levenshtein-automata/) is used.
 
 #### Document indexes
 
 The `fst` will only return the words that match with the search automaton but the goal of the search engine is to retrieve all matches in all the documents when a query is made. You want it to return some sort of position in an attribute in a document, an information about where the given word matched.
 
-To make it possible we retrieve all of the `DocIndex` corresponding to all the matching words in the fst, we use the [`WordsIndex`](https://github.com/meilisearch/MeiliDB/blob/3db823de002243004612e36a19b4578d800dab97/meilidb-data/src/database/words_index.rs#L11-L21) Tree to get the `DocIndexes` corresponding the words.
+To make it possible we retrieve all of the `DocIndex` corresponding to all the matching words in the fst, we use the [`WordsIndex`](https://github.com/meilisearch/MeiliSearch/blob/3db823de002243004612e36a19b4578d800dab97/meilisearch-data/src/database/words_index.rs#L11-L21) Tree to get the `DocIndexes` corresponding the words.
 
 ### The schema
 
-The schema is a data structure that represents which documents attributes should be stored and which should be indexed. It is stored under a the [`MainIndex`](https://github.com/meilisearch/MeiliDB/blob/3db823de002243004612e36a19b4578d800dab97/meilidb-data/src/database/main_index.rs#L12) Tree and given to MeiliDB only at the creation of an index.
+The schema is a data structure that represents which documents attributes should be stored and which should be indexed. It is stored under a the [`MainIndex`](https://github.com/meilisearch/MeiliSearch/blob/3db823de002243004612e36a19b4578d800dab97/meilisearch-data/src/database/main_index.rs#L12) Tree and given to MeiliSearch only at the creation of an index.
 
-Each document attribute is associated to a unique 16 bit number named [`SchemaAttr`](https://github.com/meilisearch/MeiliDB/blob/3db823de002243004612e36a19b4578d800dab97/meilidb-data/src/schema.rs#L186).
+Each document attribute is associated to a unique 16 bit number named [`SchemaAttr`](https://github.com/meilisearch/MeiliSearch/blob/3db823de002243004612e36a19b4578d800dab97/meilisearch-data/src/schema.rs#L186).
 
 In the future, this schema type could be given along with updates, the database could be able to handled a new schema and reindex the database according to the new one.
 
 ### Document attributes
 
-When the engine handle a query the result that the requester want is a document, not only the [`Matches`](https://github.com/meilisearch/MeiliDB/blob/3db823de002243004612e36a19b4578d800dab97/meilidb-core/src/lib.rs#L62-L88) associated to it, fields of the original document must be returned too.
+When the engine handle a query the result that the requester want is a document, not only the [`Matches`](https://github.com/meilisearch/MeiliSearch/blob/3db823de002243004612e36a19b4578d800dab97/meilisearch-core/src/lib.rs#L62-L88) associated to it, fields of the original document must be returned too.
 
-So MeiliDB again uses the power of the underlying key-value store and save the documents attributes marked as _STORE_ in the schema. The dedicated Tree for this information is the [`DocumentsIndex`](https://github.com/meilisearch/MeiliDB/blob/3db823de002243004612e36a19b4578d800dab97/meilidb-data/src/database/documents_index.rs#L11).
+So MeiliSearch again uses the power of the underlying key-value store and save the documents attributes marked as _STORE_ in the schema. The dedicated Tree for this information is the [`DocumentsIndex`](https://github.com/meilisearch/MeiliSearch/blob/3db823de002243004612e36a19b4578d800dab97/meilisearch-data/src/database/documents_index.rs#L11).
 
 When a document field is saved in the key-value store its value is binary encoded using [message pack](https://github.com/3Hren/msgpack-rust), so a document must be serializable using serde.
 
@@ -70,26 +70,26 @@ When a document field is saved in the key-value store its value is binary encode
 
 ## How is a request processed?
 
-Now that we have our inverted index we are able to return results based on a query. In the MeiliDB universe a query is a simple string containing words.
+Now that we have our inverted index we are able to return results based on a query. In the MeiliSearch universe a query is a simple string containing words.
 
 ### Query lexemes
 
-The first step to be able to call the underlying structures is to split the query in words, for that we use a [custom tokenizer](https://github.com/meilisearch/MeiliDB/blob/3db823de002243004612e36a19b4578d800dab97/meilidb-tokenizer/src/lib.rs#L82-L84). Note that a tokenizer is specialized for a human language, this is the hard part.
+The first step to be able to call the underlying structures is to split the query in words, for that we use a [custom tokenizer](https://github.com/meilisearch/MeiliSearch/blob/3db823de002243004612e36a19b4578d800dab97/meilisearch-tokenizer/src/lib.rs#L82-L84). Note that a tokenizer is specialized for a human language, this is the hard part.
 
 ### Automatons and query index
 
-So to query the fst we need an automaton, in MeiliDB we use a [levenshtein automaton](https://en.wikipedia.org/wiki/Levenshtein_automaton), this automaton is constructed using a string and a maximum distance. According to the [Algolia's blog post](https://blog.algolia.com/inside-the-algolia-engine-part-3-query-processing/#algolia%e2%80%99s-way-of-searching-for-alternatives) we [created the DFAs](https://github.com/meilisearch/MeiliDB/blob/3db823de002243004612e36a19b4578d800dab97/meilidb-core/src/automaton.rs#L59-L78) with different settings.
+So to query the fst we need an automaton, in MeiliSearch we use a [levenshtein automaton](https://en.wikipedia.org/wiki/Levenshtein_automaton), this automaton is constructed using a string and a maximum distance. According to the [Algolia's blog post](https://blog.algolia.com/inside-the-algolia-engine-part-3-query-processing/#algolia%e2%80%99s-way-of-searching-for-alternatives) we [created the DFAs](https://github.com/meilisearch/MeiliSearch/blob/3db823de002243004612e36a19b4578d800dab97/meilisearch-core/src/automaton.rs#L59-L78) with different settings.
 
 Thanks to the power of the fst library [it is possible to union multiple automatons](https://docs.rs/fst/0.3.2/fst/map/struct.OpBuilder.html#method.union) on the same fst set. The `Stream` is able to return all the matching words. We use these words to find the whole list of `DocIndexes` associated.
 
-With all these informations it is possible [to reconstruct a list of all the `DocIndexes` associated](https://github.com/meilisearch/MeiliDB/blob/3db823de002243004612e36a19b4578d800dab97/meilidb-core/src/query_builder.rs#L103-L130) with the words queried.
+With all these informations it is possible [to reconstruct a list of all the `DocIndexes` associated](https://github.com/meilisearch/MeiliSearch/blob/3db823de002243004612e36a19b4578d800dab97/meilisearch-core/src/query_builder.rs#L103-L130) with the words queried.
 
 ### Sort by criteria
 
-Now that we are able to get a big list of [DocIndexes](https://github.com/Kerollmops/MeiliDB/blob/550dc1e99224e386516877450320f694947332d4/src/lib.rs#L21-L36) it is not enough to sort them by criteria, we need more informations like the levenshtein distance or the fact that a query word match exactly the word stored in the fst. So [we stuff it a little bit](https://github.com/Kerollmops/MeiliDB/blob/550dc1e99224e386516877450320f694947332d4/src/rank/query_builder.rs#L86-L93), and aggregate all these [Matches](https://github.com/Kerollmops/MeiliDB/blob/550dc1e99224e386516877450320f694947332d4/src/lib.rs#L47-L74) for each document. This way it will be easy to sort a simple vector of document using a bunch of functions.
+Now that we are able to get a big list of [DocIndexes](https://github.com/Kerollmops/MeiliSearch/blob/550dc1e99224e386516877450320f694947332d4/src/lib.rs#L21-L36) it is not enough to sort them by criteria, we need more informations like the levenshtein distance or the fact that a query word match exactly the word stored in the fst. So [we stuff it a little bit](https://github.com/Kerollmops/MeiliSearch/blob/550dc1e99224e386516877450320f694947332d4/src/rank/query_builder.rs#L86-L93), and aggregate all these [Matches](https://github.com/Kerollmops/MeiliSearch/blob/550dc1e99224e386516877450320f694947332d4/src/lib.rs#L47-L74) for each document. This way it will be easy to sort a simple vector of document using a bunch of functions.
 
-With this big list of documents and associated matches [we are able to sort only the part of the slice that we want](https://github.com/meilisearch/MeiliDB/blob/3db823de002243004612e36a19b4578d800dab97/meilidb-core/src/query_builder.rs#L160-L188) using bucket sorting. [Each criterion](https://github.com/meilisearch/MeiliDB/blob/3db823de002243004612e36a19b4578d800dab97/meilidb-core/src/criterion/mod.rs#L95-L101) is evaluated on each subslice without copy, thanks to [GroupByMut](https://docs.rs/slice-group-by/0.2.4/slice_group_by/) which, I hope [will soon be merged](https://github.com/rust-lang/rfcs/pull/2477).
+With this big list of documents and associated matches [we are able to sort only the part of the slice that we want](https://github.com/meilisearch/MeiliSearch/blob/3db823de002243004612e36a19b4578d800dab97/meilisearch-core/src/query_builder.rs#L160-L188) using bucket sorting. [Each criterion](https://github.com/meilisearch/MeiliSearch/blob/3db823de002243004612e36a19b4578d800dab97/meilisearch-core/src/criterion/mod.rs#L95-L101) is evaluated on each subslice without copy, thanks to [GroupByMut](https://docs.rs/slice-group-by/0.2.4/slice_group_by/) which, I hope [will soon be merged](https://github.com/rust-lang/rfcs/pull/2477).
 
-Note that it is possible to customize the criteria used by using the `QueryBuilder::with_criteria` constructor, this way you can implement some custom ranking based on the document attributes using the appropriate structure and the [`document` method](https://github.com/meilisearch/MeiliDB/blob/3db823de002243004612e36a19b4578d800dab97/meilidb-data/src/database/index.rs#L86).
+Note that it is possible to customize the criteria used by using the `QueryBuilder::with_criteria` constructor, this way you can implement some custom ranking based on the document attributes using the appropriate structure and the [`document` method](https://github.com/meilisearch/MeiliSearch/blob/3db823de002243004612e36a19b4578d800dab97/meilisearch-data/src/database/index.rs#L86).
 
-At this point, MeiliDB work is over 🎉
+At this point, MeiliSearch work is over 🎉

meilidb-core/src/database.rs

@@ -1,205 +0,0 @@
-use std::collections::hash_map::{Entry, HashMap};
-use std::fs::File;
-use std::path::Path;
-use std::sync::{Arc, RwLock};
-use std::{fs, thread};
-
-use crossbeam_channel::Receiver;
-use heed::types::{Str, Unit};
-use heed::{CompactionOption, Result as ZResult};
-use log::{debug, error};
-
-use crate::{store, update, Index, MResult};
-
-pub type BoxUpdateFn = Box<dyn Fn(update::ProcessedUpdateResult) + Send + Sync + 'static>;
-type ArcSwapFn = arc_swap::ArcSwapOption<BoxUpdateFn>;
-
-pub struct Database {
-    pub env: heed::Env,
-    common_store: heed::PolyDatabase,
-    indexes_store: heed::Database<Str, Unit>,
-    indexes: RwLock<HashMap<String, (Index, Arc<ArcSwapFn>, thread::JoinHandle<()>)>>,
-}
-
-fn update_awaiter(receiver: Receiver<()>, env: heed::Env, update_fn: Arc<ArcSwapFn>, index: Index) {
-    for () in receiver {
-        // consume all updates in order (oldest first)
-        loop {
-            let mut writer = match env.write_txn() {
-                Ok(writer) => writer,
-                Err(e) => {
-                    error!("LMDB writer transaction begin failed: {}", e);
-                    break;
-                }
-            };
-
-            match update::update_task(&mut writer, index.clone()) {
-                Ok(Some(status)) => {
-                    match status.result {
-                        Ok(_) => {
-                            if let Err(e) = writer.commit() {
-                                error!("update transaction failed: {}", e)
-                            }
-                        }
-                        Err(_) => writer.abort(),
-                    }
-
-                    if let Some(ref callback) = *update_fn.load() {
-                        (callback)(status);
-                    }
-                }
-                // no more updates to handle for now
-                Ok(None) => {
-                    debug!("no more updates");
-                    writer.abort();
-                    break;
-                }
-                Err(e) => {
-                    error!("update task failed: {}", e);
-                    writer.abort()
-                }
-            }
-        }
-    }
-}
-
-impl Database {
-    pub fn open_or_create(path: impl AsRef<Path>) -> MResult<Database> {
-        fs::create_dir_all(path.as_ref())?;
-
-        let env = heed::EnvOpenOptions::new()
-            .map_size(10 * 1024 * 1024 * 1024) // 10GB
-            .max_dbs(3000)
-            .open(path)?;
-
-        let common_store = env.create_dyn_database(Some("common"))?;
-        let indexes_store = env.create_database::<Str, Unit>(Some("indexes"))?;
-
-        // list all indexes that needs to be opened
-        let mut must_open = Vec::new();
-        let reader = env.read_txn()?;
-        for result in indexes_store.iter(&reader)? {
-            let (index_name, _) = result?;
-            must_open.push(index_name.to_owned());
-        }
-
-        reader.abort();
-
-        // open the previously aggregated indexes
-        let mut indexes = HashMap::new();
-        for index_name in must_open {
-            let (sender, receiver) = crossbeam_channel::bounded(100);
-            let index = match store::open(&env, &index_name, sender.clone())? {
-                Some(index) => index,
-                None => {
-                    log::warn!(
-                        "the index {} doesn't exist or has not all the databases",
-                        index_name
-                    );
-                    continue;
-                }
-            };
-            let update_fn = Arc::new(ArcSwapFn::empty());
-
-            let env_clone = env.clone();
-            let index_clone = index.clone();
-            let update_fn_clone = update_fn.clone();
-
-            let handle = thread::spawn(move || {
-                update_awaiter(receiver, env_clone, update_fn_clone, index_clone)
-            });
-
-            // send an update notification to make sure that
-            // possible pre-boot updates are consumed
-            sender.send(()).unwrap();
-
-            let result = indexes.insert(index_name, (index, update_fn, handle));
-            assert!(
-                result.is_none(),
-                "The index should not have been already open"
-            );
-        }
-
-        Ok(Database {
-            env,
-            common_store,
-            indexes_store,
-            indexes: RwLock::new(indexes),
-        })
-    }
-
-    pub fn open_index(&self, name: impl AsRef<str>) -> Option<Index> {
-        let indexes_lock = self.indexes.read().unwrap();
-        match indexes_lock.get(name.as_ref()) {
-            Some((index, ..)) => Some(index.clone()),
-            None => None,
-        }
-    }
-
-    pub fn create_index(&self, name: impl AsRef<str>) -> MResult<Index> {
-        let name = name.as_ref();
-        let mut indexes_lock = self.indexes.write().unwrap();
-
-        match indexes_lock.entry(name.to_owned()) {
-            Entry::Occupied(_) => Err(crate::Error::IndexAlreadyExists),
-            Entry::Vacant(entry) => {
-                let (sender, receiver) = crossbeam_channel::bounded(100);
-                let index = store::create(&self.env, name, sender)?;
-
-                let mut writer = self.env.write_txn()?;
-                self.indexes_store.put(&mut writer, name, &())?;
-
-                let env_clone = self.env.clone();
-                let index_clone = index.clone();
-
-                let no_update_fn = Arc::new(ArcSwapFn::empty());
-                let no_update_fn_clone = no_update_fn.clone();
-
-                let handle = thread::spawn(move || {
-                    update_awaiter(receiver, env_clone, no_update_fn_clone, index_clone)
-                });
-
-                writer.commit()?;
-                entry.insert((index.clone(), no_update_fn, handle));
-
-                Ok(index)
-            }
-        }
-    }
-
-    pub fn set_update_callback(&self, name: impl AsRef<str>, update_fn: BoxUpdateFn) -> bool {
-        let indexes_lock = self.indexes.read().unwrap();
-        match indexes_lock.get(name.as_ref()) {
-            Some((_, current_update_fn, _)) => {
-                let update_fn = Some(Arc::new(update_fn));
-                current_update_fn.swap(update_fn);
-                true
-            }
-            None => false,
-        }
-    }
-
-    pub fn unset_update_callback(&self, name: impl AsRef<str>) -> bool {
-        let indexes_lock = self.indexes.read().unwrap();
-        match indexes_lock.get(name.as_ref()) {
-            Some((_, current_update_fn, _)) => {
-                current_update_fn.swap(None);
-                true
-            }
-            None => false,
-        }
-    }
-
-    pub fn copy_and_compact_to_path<P: AsRef<Path>>(&self, path: P) -> ZResult<File> {
-        self.env.copy_to_path(path, CompactionOption::Enabled)
-    }
-
-    pub fn indexes_names(&self) -> MResult<Vec<String>> {
-        let indexes = self.indexes.read().unwrap();
-        Ok(indexes.keys().cloned().collect())
-    }
-
-    pub fn common_store(&self) -> heed::PolyDatabase {
-        self.common_store
-    }
-}

meilidb-core/src/lib.rs

@@ -1,169 +0,0 @@
-#[cfg(test)]
-#[macro_use]
-extern crate assert_matches;
-
-mod automaton;
-pub mod criterion;
-mod database;
-mod distinct_map;
-mod error;
-mod levenshtein;
-mod number;
-mod query_builder;
-mod ranked_map;
-mod raw_document;
-pub mod raw_indexer;
-mod reordered_attrs;
-pub mod serde;
-pub mod store;
-mod update;
-
-pub use self::database::{BoxUpdateFn, Database};
-pub use self::error::{Error, MResult};
-pub use self::number::{Number, ParseNumberError};
-pub use self::ranked_map::RankedMap;
-pub use self::raw_document::RawDocument;
-pub use self::store::Index;
-pub use self::update::{EnqueuedUpdateResult, ProcessedUpdateResult, UpdateStatus, UpdateType};
-
-use ::serde::{Deserialize, Serialize};
-use zerocopy::{AsBytes, FromBytes};
-
-/// Represent an internally generated document unique identifier.
-///
-/// It is used to inform the database the document you want to deserialize.
-/// Helpful for custom ranking.
-#[derive(
-    Debug,
-    Copy,
-    Clone,
-    Eq,
-    PartialEq,
-    PartialOrd,
-    Ord,
-    Hash,
-    Serialize,
-    Deserialize,
-    AsBytes,
-    FromBytes,
-)]
-#[repr(C)]
-pub struct DocumentId(pub u64);
-
-/// This structure represent the position of a word
-/// in a document and its attributes.
-///
-/// This is stored in the map, generated at index time,
-/// extracted and interpreted at search time.
-#[derive(Debug, Copy, Clone, PartialEq, Eq, PartialOrd, Ord, Hash, AsBytes, FromBytes)]
-#[repr(C)]
-pub struct DocIndex {
-    /// The document identifier where the word was found.
-    pub document_id: DocumentId,
-
-    /// The attribute in the document where the word was found
-    /// along with the index in it.
-    pub attribute: u16,
-    pub word_index: u16,
-
-    /// The position in bytes where the word was found
-    /// along with the length of it.
-    ///
-    /// It informs on the original word area in the text indexed
-    /// without needing to run the tokenizer again.
-    pub char_index: u16,
-    pub char_length: u16,
-}
-
-/// This structure represent a matching word with informations
-/// on the location of the word in the document.
-///
-/// The order of the field is important because it defines
-/// the way these structures are ordered between themselves.
-#[derive(Debug, Copy, Clone, PartialEq, Eq, PartialOrd, Ord, Hash)]
-pub struct Highlight {
-    /// The attribute in the document where the word was found
-    /// along with the index in it.
-    pub attribute: u16,
-
-    /// The position in bytes where the word was found.
-    ///
-    /// It informs on the original word area in the text indexed
-    /// without needing to run the tokenizer again.
-    pub char_index: u16,
-
-    /// The length in bytes of the found word.
-    ///
-    /// It informs on the original word area in the text indexed
-    /// without needing to run the tokenizer again.
-    pub char_length: u16,
-}
-
-#[doc(hidden)]
-#[derive(Debug, Copy, Clone, PartialEq, Eq, PartialOrd, Ord, Hash)]
-pub struct TmpMatch {
-    pub query_index: u32,
-    pub distance: u8,
-    pub attribute: u16,
-    pub word_index: u16,
-    pub is_exact: bool,
-}
-
-#[derive(Debug, Clone, PartialEq, Eq, PartialOrd, Ord, Hash)]
-pub struct Document {
-    pub id: DocumentId,
-    pub highlights: Vec<Highlight>,
-
-    #[cfg(test)]
-    pub matches: Vec<TmpMatch>,
-}
-
-impl Document {
-    #[cfg(not(test))]
-    fn from_raw(raw: RawDocument) -> Document {
-        Document {
-            id: raw.id,
-            highlights: raw.highlights,
-        }
-    }
-
-    #[cfg(test)]
-    fn from_raw(raw: RawDocument) -> Document {
-        let len = raw.query_index().len();
-        let mut matches = Vec::with_capacity(len);
-
-        let query_index = raw.query_index();
-        let distance = raw.distance();
-        let attribute = raw.attribute();
-        let word_index = raw.word_index();
-        let is_exact = raw.is_exact();
-
-        for i in 0..len {
-            let match_ = TmpMatch {
-                query_index: query_index[i],
-                distance: distance[i],
-                attribute: attribute[i],
-                word_index: word_index[i],
-                is_exact: is_exact[i],
-            };
-            matches.push(match_);
-        }
-
-        Document {
-            id: raw.id,
-            matches,
-            highlights: raw.highlights,
-        }
-    }
-}
-
-#[cfg(test)]
-mod tests {
-    use super::*;
-    use std::mem;
-
-    #[test]
-    fn docindex_mem_size() {
-        assert_eq!(mem::size_of::<DocIndex>(), 16);
-    }
-}

meilidb-core/src/store/main.rs

@@ -1,121 +0,0 @@
-use crate::RankedMap;
-use heed::types::{ByteSlice, OwnedType, SerdeBincode, Str};
-use heed::Result as ZResult;
-use meilidb_schema::Schema;
-use std::sync::Arc;
-
-const CUSTOMS_KEY: &str = "customs-key";
-const NUMBER_OF_DOCUMENTS_KEY: &str = "number-of-documents";
-const RANKED_MAP_KEY: &str = "ranked-map";
-const SCHEMA_KEY: &str = "schema";
-const SYNONYMS_KEY: &str = "synonyms";
-const STOP_WORDS_KEY: &str = "stop-words";
-const WORDS_KEY: &str = "words";
-
-#[derive(Copy, Clone)]
-pub struct Main {
-    pub(crate) main: heed::PolyDatabase,
-}
-
-impl Main {
-    pub fn put_words_fst(self, writer: &mut heed::RwTxn, fst: &fst::Set) -> ZResult<()> {
-        let bytes = fst.as_fst().as_bytes();
-        self.main.put::<Str, ByteSlice>(writer, WORDS_KEY, bytes)
-    }
-
-    pub fn words_fst(self, reader: &heed::RoTxn) -> ZResult<Option<fst::Set>> {
-        match self.main.get::<Str, ByteSlice>(reader, WORDS_KEY)? {
-            Some(bytes) => {
-                let len = bytes.len();
-                let bytes = Arc::from(bytes);
-                let fst = fst::raw::Fst::from_shared_bytes(bytes, 0, len).unwrap();
-                Ok(Some(fst::Set::from(fst)))
-            }
-            None => Ok(None),
-        }
-    }
-
-    pub fn put_schema(self, writer: &mut heed::RwTxn, schema: &Schema) -> ZResult<()> {
-        self.main
-            .put::<Str, SerdeBincode<Schema>>(writer, SCHEMA_KEY, schema)
-    }
-
-    pub fn schema(self, reader: &heed::RoTxn) -> ZResult<Option<Schema>> {
-        self.main
-            .get::<Str, SerdeBincode<Schema>>(reader, SCHEMA_KEY)
-    }
-
-    pub fn put_ranked_map(self, writer: &mut heed::RwTxn, ranked_map: &RankedMap) -> ZResult<()> {
-        self.main
-            .put::<Str, SerdeBincode<RankedMap>>(writer, RANKED_MAP_KEY, &ranked_map)
-    }
-
-    pub fn ranked_map(self, reader: &heed::RoTxn) -> ZResult<Option<RankedMap>> {
-        self.main
-            .get::<Str, SerdeBincode<RankedMap>>(reader, RANKED_MAP_KEY)
-    }
-
-    pub fn put_synonyms_fst(self, writer: &mut heed::RwTxn, fst: &fst::Set) -> ZResult<()> {
-        let bytes = fst.as_fst().as_bytes();
-        self.main.put::<Str, ByteSlice>(writer, SYNONYMS_KEY, bytes)
-    }
-
-    pub fn synonyms_fst(self, reader: &heed::RoTxn) -> ZResult<Option<fst::Set>> {
-        match self.main.get::<Str, ByteSlice>(reader, SYNONYMS_KEY)? {
-            Some(bytes) => {
-                let len = bytes.len();
-                let bytes = Arc::from(bytes);
-                let fst = fst::raw::Fst::from_shared_bytes(bytes, 0, len).unwrap();
-                Ok(Some(fst::Set::from(fst)))
-            }
-            None => Ok(None),
-        }
-    }
-
-    pub fn put_stop_words_fst(self, writer: &mut heed::RwTxn, fst: &fst::Set) -> ZResult<()> {
-        let bytes = fst.as_fst().as_bytes();
-        self.main
-            .put::<Str, ByteSlice>(writer, STOP_WORDS_KEY, bytes)
-    }
-
-    pub fn stop_words_fst(self, reader: &heed::RoTxn) -> ZResult<Option<fst::Set>> {
-        match self.main.get::<Str, ByteSlice>(reader, STOP_WORDS_KEY)? {
-            Some(bytes) => {
-                let len = bytes.len();
-                let bytes = Arc::from(bytes);
-                let fst = fst::raw::Fst::from_shared_bytes(bytes, 0, len).unwrap();
-                Ok(Some(fst::Set::from(fst)))
-            }
-            None => Ok(None),
-        }
-    }
-
-    pub fn put_number_of_documents<F>(self, writer: &mut heed::RwTxn, f: F) -> ZResult<u64>
-    where
-        F: Fn(u64) -> u64,
-    {
-        let new = self.number_of_documents(writer).map(f)?;
-        self.main
-            .put::<Str, OwnedType<u64>>(writer, NUMBER_OF_DOCUMENTS_KEY, &new)?;
-        Ok(new)
-    }
-
-    pub fn number_of_documents(self, reader: &heed::RoTxn) -> ZResult<u64> {
-        match self
-            .main
-            .get::<Str, OwnedType<u64>>(reader, NUMBER_OF_DOCUMENTS_KEY)?
-        {
-            Some(value) => Ok(value),
-            None => Ok(0),
-        }
-    }
-
-    pub fn put_customs(self, writer: &mut heed::RwTxn, customs: &[u8]) -> ZResult<()> {
-        self.main
-            .put::<Str, ByteSlice>(writer, CUSTOMS_KEY, customs)
-    }
-
-    pub fn customs<'txn>(self, reader: &'txn heed::RoTxn) -> ZResult<Option<&'txn [u8]>> {
-        self.main.get::<Str, ByteSlice>(reader, CUSTOMS_KEY)
-    }
-}

meilisearch-core/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
-name = "meilidb-core"
-version = "0.5.11"
+name = "meilisearch-core"
+version = "0.8.1"
 authors = ["Kerollmops <clement@meilisearch.com>"]
 edition = "2018"
 
@@ -8,32 +8,27 @@ edition = "2018"
 arc-swap = "0.4.3"
 bincode = "1.1.4"
 byteorder = "1.3.2"
-crossbeam-channel = "0.3.9"
+chrono = { version = "0.4.9", features = ["serde"] }
+crossbeam-channel = "0.4.0"
 deunicode = "1.0.0"
 env_logger = "0.7.0"
+fst = { version = "0.3.5", default-features = false }
 hashbrown = { version = "0.6.0", features = ["serde"] }
-heed = "0.1.0"
+heed = "0.6.0"
+levenshtein_automata = { version = "0.1.1", features = ["fst_automaton"] }
 log = "0.4.8"
-meilidb-schema = { path = "../meilidb-schema", version = "0.5.11" }
-meilidb-tokenizer = { path = "../meilidb-tokenizer", version = "0.5.11" }
+meilisearch-schema = { path = "../meilisearch-schema", version = "0.8.1" }
+meilisearch-tokenizer = { path = "../meilisearch-tokenizer", version = "0.8.1" }
+meilisearch-types = { path = "../meilisearch-types", version = "0.8.1" }
 once_cell = "1.2.0"
 ordered-float = { version = "1.0.2", features = ["serde"] }
 sdset = "0.3.3"
 serde = { version = "1.0.101", features = ["derive"] }
 serde_json = "1.0.41"
-siphasher = "0.3.0"
+siphasher = "0.3.1"
 slice-group-by = "0.2.6"
 zerocopy = "0.2.8"
 
-[dependencies.levenshtein_automata]
-git = "https://github.com/Kerollmops/levenshtein-automata.git"
-branch = "arc-byte-slice"
-features = ["fst_automaton"]
-
-[dependencies.fst]
-git = "https://github.com/Kerollmops/fst.git"
-branch = "arc-byte-slice"
-
 [dev-dependencies]
 assert_matches = "1.3"
 csv = "1.0.7"

meilisearch-core/examples/from_file.rs

@@ -12,10 +12,8 @@ use serde::{Deserialize, Serialize};
 use structopt::StructOpt;
 use termcolor::{Color, ColorChoice, ColorSpec, StandardStream, WriteColor};
 
-use meilidb_core::{Database, Highlight, ProcessedUpdateResult};
-use meilidb_schema::SchemaAttr;
-
-const INDEX_NAME: &str = "default";
+use meilisearch_core::{Database, Highlight, ProcessedUpdateResult};
+use meilisearch_schema::SchemaAttr;
 
 #[derive(Debug, StructOpt)]
 struct IndexCommand {
@@ -23,6 +21,9 @@ struct IndexCommand {
     #[structopt(parse(from_os_str))]
     database_path: PathBuf,
 
+    #[structopt(long, default_value = "default")]
+    index_uid: String,
+
     /// The csv file to index.
     #[structopt(parse(from_os_str))]
     csv_data_path: PathBuf,
@@ -44,6 +45,9 @@ struct SearchCommand {
     #[structopt(parse(from_os_str))]
     database_path: PathBuf,
 
+    #[structopt(long, default_value = "default")]
+    index_uid: String,
+
     /// Timeout after which the search will return results.
     #[structopt(long)]
     fetch_timeout_ms: Option<u64>,
@@ -70,6 +74,9 @@ struct ShowUpdatesCommand {
     /// The path of the database to work with.
     #[structopt(parse(from_os_str))]
     database_path: PathBuf,
+
+    #[structopt(long, default_value = "default")]
+    index_uid: String,
 }
 
 #[derive(Debug, StructOpt)]
@@ -97,33 +104,34 @@ fn index_command(command: IndexCommand, database: Database) -> Result<(), Box<dy
     let start = Instant::now();
 
     let (sender, receiver) = mpsc::sync_channel(100);
-    let update_fn = move |update: ProcessedUpdateResult| sender.send(update.update_id).unwrap();
-    let index = match database.open_index(INDEX_NAME) {
+    let update_fn =
+        move |_name: &str, update: ProcessedUpdateResult| sender.send(update.update_id).unwrap();
+    let index = match database.open_index(&command.index_uid) {
         Some(index) => index,
-        None => database.create_index(INDEX_NAME).unwrap(),
+        None => database.create_index(&command.index_uid).unwrap(),
     };
 
-    let done = database.set_update_callback(INDEX_NAME, Box::new(update_fn));
-    assert!(done, "could not set the index update function");
+    database.set_update_callback(Box::new(update_fn));
 
-    let env = &database.env;
+    let db = &database;
 
     let schema = {
         let string = fs::read_to_string(&command.schema)?;
         toml::from_str(&string).unwrap()
     };
 
-    let mut writer = env.write_txn().unwrap();
-    match index.main.schema(&writer)? {
+    let reader = db.main_read_txn().unwrap();
+    let mut update_writer = db.update_write_txn().unwrap();
+    match index.main.schema(&reader)? {
         Some(current_schema) => {
             if current_schema != schema {
-                return Err(meilidb_core::Error::SchemaDiffer.into());
+                return Err(meilisearch_core::Error::SchemaDiffer.into());
             }
-            writer.abort();
+            update_writer.abort();
         }
         None => {
-            index.schema_update(&mut writer, schema)?;
-            writer.commit().unwrap();
+            index.schema_update(&mut update_writer, schema)?;
+            update_writer.commit().unwrap();
         }
     }
 
@@ -166,10 +174,10 @@ fn index_command(command: IndexCommand, database: Database) -> Result<(), Box<dy
 
         println!();
 
-        let mut writer = env.write_txn().unwrap();
+        let mut update_writer = db.update_write_txn().unwrap();
         println!("committing update...");
-        let update_id = additions.finalize(&mut writer)?;
-        writer.commit().unwrap();
+        let update_id = additions.finalize(&mut update_writer)?;
+        update_writer.commit().unwrap();
         max_update_id = max_update_id.max(update_id);
         println!("committed update {}", update_id);
     }
@@ -188,8 +196,9 @@ fn index_command(command: IndexCommand, database: Database) -> Result<(), Box<dy
     );
 
     if let Some(path) = command.compact_to_path {
+        fs::create_dir_all(&path)?;
         let start = Instant::now();
-        let _file = database.copy_and_compact_to_path(&path)?;
+        let _file = database.copy_and_compact_to_path(path.join("data.mdb"))?;
         println!(
             "database compacted in {:.2?} at: {:?}",
             start.elapsed(),
@@ -210,7 +219,11 @@ fn display_highlights(text: &str, ranges: &[usize]) -> io::Result<()> {
             _ => unreachable!(),
         };
         if highlighted {
-            stdout.set_color(ColorSpec::new().set_fg(Some(Color::Yellow)))?;
+            stdout.set_color(
+                ColorSpec::new()
+                    .set_fg(Some(Color::Yellow))
+                    .set_underline(true),
+            )?;
         }
         write!(&mut stdout, "{}", &text[start..end])?;
         stdout.reset()?;
@@ -304,16 +317,16 @@ fn crop_text(
 }
 
 fn search_command(command: SearchCommand, database: Database) -> Result<(), Box<dyn Error>> {
-    let env = &database.env;
+    let db = &database;
     let index = database
-        .open_index(INDEX_NAME)
+        .open_index(&command.index_uid)
         .expect("Could not find index");
 
-    let reader = env.read_txn().unwrap();
+    let reader = db.main_read_txn().unwrap();
     let schema = index.main.schema(&reader)?;
     reader.abort();
 
-    let schema = schema.ok_or(meilidb_core::Error::SchemaMissing)?;
+    let schema = schema.ok_or(meilisearch_core::Error::SchemaMissing)?;
 
     let fields = command.displayed_fields.iter().map(String::as_str);
     let fields = HashSet::from_iter(fields);
@@ -327,7 +340,7 @@ fn search_command(command: SearchCommand, database: Database) -> Result<(), Box<
             Ok(query) => {
                 let start_total = Instant::now();
 
-                let reader = env.read_txn().unwrap();
+                let reader = db.main_read_txn().unwrap();
                 let ref_index = &index;
                 let ref_reader = &reader;
 
@@ -429,15 +442,15 @@ fn search_command(command: SearchCommand, database: Database) -> Result<(), Box<
 }
 
 fn show_updates_command(
-    _command: ShowUpdatesCommand,
+    command: ShowUpdatesCommand,
     database: Database,
 ) -> Result<(), Box<dyn Error>> {
-    let env = &database.env;
+    let db = &database;
     let index = database
-        .open_index(INDEX_NAME)
+        .open_index(&command.index_uid)
         .expect("Could not find index");
 
-    let reader = env.read_txn().unwrap();
+    let reader = db.update_read_txn().unwrap();
     let updates = index.all_updates_status(&reader)?;
     println!("{:#?}", updates);
     reader.abort();

meilisearch-core/src/automaton/mod.rs

@@ -6,8 +6,9 @@ use std::{cmp, vec};
 
 use fst::{IntoStreamer, Streamer};
 use levenshtein_automata::DFA;
-use meilidb_tokenizer::{is_cjk, split_query_string};
+use meilisearch_tokenizer::{is_cjk, split_query_string};
 
+use crate::database::MainT;
 use crate::error::MResult;
 use crate::store;
 
@@ -23,7 +24,7 @@ pub struct AutomatonProducer {
 
 impl AutomatonProducer {
     pub fn new(
-        reader: &heed::RoTxn,
+        reader: &heed::RoTxn<MainT>,
         query: &str,
         main_store: store::Main,
         postings_list_store: store::PostingsLists,
@@ -131,7 +132,7 @@ pub fn normalize_str(string: &str) -> String {
 }
 
 fn split_best_frequency<'a>(
-    reader: &heed::RoTxn,
+    reader: &heed::RoTxn<MainT>,
     word: &'a str,
     postings_lists_store: store::PostingsLists,
 ) -> MResult<Option<(&'a str, &'a str)>> {
@@ -159,7 +160,7 @@ fn split_best_frequency<'a>(
 }
 
 fn generate_automatons(
-    reader: &heed::RoTxn,
+    reader: &heed::RoTxn<MainT>,
     query: &str,
     main_store: store::Main,
     postings_lists_store: store::PostingsLists,

meilisearch-core/src/criterion/exact.rs

@@ -1,6 +1,6 @@
 use std::cmp::Ordering;
 
-use meilidb_schema::SchemaAttr;
+use meilisearch_schema::SchemaAttr;
 use sdset::Set;
 use slice_group_by::GroupBy;
 

meilisearch-core/src/criterion/sort_by_attr.rs

@@ -4,7 +4,7 @@ use std::fmt;
 
 use crate::criterion::Criterion;
 use crate::{RankedMap, RawDocument};
-use meilidb_schema::{Schema, SchemaAttr};
+use meilisearch_schema::{Schema, SchemaAttr};
 
 /// An helper struct that permit to sort documents by
 /// some of their stored attributes.
@@ -23,7 +23,7 @@ use meilidb_schema::{Schema, SchemaAttr};
 ///
 /// ```ignore
 /// use serde_derive::Deserialize;
-/// use meilidb::rank::criterion::*;
+/// use meilisearch::rank::criterion::*;
 ///
 /// let custom_ranking = SortByAttr::lower_is_better(&ranked_map, &schema, "published_at")?;
 ///

meilisearch-core/src/error.rs

@@ -97,7 +97,7 @@ pub enum UnsupportedOperation {
     SchemaAlreadyExists,
     CannotUpdateSchemaIdentifier,
     CannotReorderSchemaAttribute,
-    CannotIntroduceNewSchemaAttribute,
+    CanOnlyIntroduceNewSchemaAttributesAtEnd,
     CannotRemoveSchemaAttribute,
 }
 
@@ -108,8 +108,8 @@ impl fmt::Display for UnsupportedOperation {
             SchemaAlreadyExists => write!(f, "Cannot update index which already have a schema"),
             CannotUpdateSchemaIdentifier => write!(f, "Cannot update the identifier of a schema"),
             CannotReorderSchemaAttribute => write!(f, "Cannot reorder the attributes of a schema"),
-            CannotIntroduceNewSchemaAttribute => {
-                write!(f, "Cannot introduce new attributes in a schema")
+            CanOnlyIntroduceNewSchemaAttributesAtEnd => {
+                write!(f, "Can only introduce new attributes at end of a schema")
             }
             CannotRemoveSchemaAttribute => write!(f, "Cannot remove attributes from a schema"),
         }

meilisearch-core/src/lib.rs

@@ -0,0 +1,97 @@
+#[cfg(test)]
+#[macro_use]
+extern crate assert_matches;
+
+mod automaton;
+pub mod criterion;
+mod database;
+mod distinct_map;
+mod error;
+mod levenshtein;
+mod number;
+mod query_builder;
+mod ranked_map;
+mod raw_document;
+pub mod raw_indexer;
+mod reordered_attrs;
+pub mod serde;
+pub mod store;
+mod update;
+
+pub use self::database::{BoxUpdateFn, Database, MainT, UpdateT};
+pub use self::error::{Error, MResult};
+pub use self::number::{Number, ParseNumberError};
+pub use self::ranked_map::RankedMap;
+pub use self::raw_document::RawDocument;
+pub use self::store::Index;
+pub use self::update::{EnqueuedUpdateResult, ProcessedUpdateResult, UpdateStatus, UpdateType};
+pub use meilisearch_types::{DocIndex, DocumentId, Highlight};
+
+#[doc(hidden)]
+#[derive(Debug, Copy, Clone, PartialEq, Eq, PartialOrd, Ord, Hash)]
+pub struct TmpMatch {
+    pub query_index: u32,
+    pub distance: u8,
+    pub attribute: u16,
+    pub word_index: u16,
+    pub is_exact: bool,
+}
+
+#[derive(Debug, Clone, PartialEq, Eq, PartialOrd, Ord, Hash)]
+pub struct Document {
+    pub id: DocumentId,
+    pub highlights: Vec<Highlight>,
+
+    #[cfg(test)]
+    pub matches: Vec<TmpMatch>,
+}
+
+impl Document {
+    #[cfg(not(test))]
+    fn from_raw(raw: RawDocument) -> Document {
+        Document {
+            id: raw.id,
+            highlights: raw.highlights,
+        }
+    }
+
+    #[cfg(test)]
+    fn from_raw(raw: RawDocument) -> Document {
+        let len = raw.query_index().len();
+        let mut matches = Vec::with_capacity(len);
+
+        let query_index = raw.query_index();
+        let distance = raw.distance();
+        let attribute = raw.attribute();
+        let word_index = raw.word_index();
+        let is_exact = raw.is_exact();
+
+        for i in 0..len {
+            let match_ = TmpMatch {
+                query_index: query_index[i],
+                distance: distance[i],
+                attribute: attribute[i],
+                word_index: word_index[i],
+                is_exact: is_exact[i],
+            };
+            matches.push(match_);
+        }
+
+        Document {
+            id: raw.id,
+            matches,
+            highlights: raw.highlights,
+        }
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+    use std::mem;
+
+    #[test]
+    fn docindex_mem_size() {
+        assert_eq!(mem::size_of::<DocIndex>(), 16);
+    }
+}

meilisearch-core/src/number.rs

@@ -1,3 +1,4 @@
+use std::cmp::Ordering;
 use std::fmt;
 use std::num::{ParseFloatError, ParseIntError};
 use std::str::FromStr;
@@ -5,7 +6,7 @@ use std::str::FromStr;
 use ordered_float::OrderedFloat;
 use serde::{Deserialize, Serialize};
 
-#[derive(Serialize, Deserialize, Debug, Copy, Clone, PartialEq, Eq, PartialOrd, Ord, Hash)]
+#[derive(Serialize, Deserialize, Debug, Copy, Clone, Hash)]
 pub enum Number {
     Unsigned(u64),
     Signed(i64),
@@ -39,6 +40,50 @@ impl FromStr for Number {
     }
 }
 
+impl PartialEq for Number {
+    fn eq(&self, other: &Number) -> bool {
+        self.cmp(other) == Ordering::Equal
+    }
+}
+
+impl Eq for Number {}
+
+impl PartialOrd for Number {
+    fn partial_cmp(&self, other: &Number) -> Option<Ordering> {
+        Some(self.cmp(other))
+    }
+}
+
+impl Ord for Number {
+    fn cmp(&self, other: &Self) -> Ordering {
+        use Number::{Float, Signed, Unsigned};
+
+        match (*self, *other) {
+            (Unsigned(a), Unsigned(b)) => a.cmp(&b),
+            (Unsigned(a), Signed(b)) => {
+                if b < 0 {
+                    Ordering::Greater
+                } else {
+                    a.cmp(&(b as u64))
+                }
+            }
+            (Unsigned(a), Float(b)) => (OrderedFloat(a as f64)).cmp(&b),
+            (Signed(a), Unsigned(b)) => {
+                if a < 0 {
+                    Ordering::Less
+                } else {
+                    (a as u64).cmp(&b)
+                }
+            }
+            (Signed(a), Signed(b)) => a.cmp(&b),
+            (Signed(a), Float(b)) => OrderedFloat(a as f64).cmp(&b),
+            (Float(a), Unsigned(b)) => a.cmp(&OrderedFloat(b as f64)),
+            (Float(a), Signed(b)) => a.cmp(&OrderedFloat(b as f64)),
+            (Float(a), Float(b)) => a.cmp(&b),
+        }
+    }
+}
+
 #[derive(Debug, Clone, PartialEq, Eq)]
 pub struct ParseNumberError {
     uint_error: ParseIntError,

meilisearch-core/src/query_builder.rs

@@ -1,14 +1,15 @@
 use hashbrown::HashMap;
 use std::convert::TryFrom;
-use std::mem;
 use std::ops::Range;
 use std::rc::Rc;
 use std::time::{Duration, Instant};
+use std::{cmp, mem};
 
 use fst::{IntoStreamer, Streamer};
 use sdset::SetBuf;
 use slice_group_by::{GroupBy, GroupByMut};
 
+use crate::database::MainT;
 use crate::automaton::{Automaton, AutomatonGroup, AutomatonProducer, QueryEnhancer};
 use crate::distinct_map::{BufferedDistinctMap, DistinctMap};
 use crate::levenshtein::prefix_damerau_levenshtein;
@@ -139,7 +140,7 @@ fn multiword_rewrite_matches(
 }
 
 fn fetch_raw_documents(
-    reader: &heed::RoTxn,
+    reader: &heed::RoTxn<MainT>,
     automatons_groups: &[AutomatonGroup],
     query_enhancer: &QueryEnhancer,
     searchables: Option<&ReorderedAttrs>,
@@ -178,7 +179,7 @@ fn fetch_raw_documents(
                 let distance = dfa.eval(input).to_u8();
                 let is_exact = *is_exact && distance == 0 && input.len() == *query_len;
 
-                let covered_area = if query.len() > input.len() {
+                let covered_area = if *query_len > input.len() {
                     input.len()
                 } else {
                     prefix_damerau_levenshtein(query.as_bytes(), input).1
@@ -202,10 +203,13 @@ fn fetch_raw_documents(
                             is_exact,
                         };
 
+                        let covered_area = u16::try_from(covered_area).unwrap_or(u16::max_value());
+                        let covered_area = cmp::min(covered_area, di.char_length);
+
                         let highlight = Highlight {
                             attribute: di.attribute,
                             char_index: di.char_index,
-                            char_length: u16::try_from(covered_area).unwrap_or(u16::max_value()),
+                            char_length: covered_area,
                         };
 
                         tmp_matches.push((di.document_id, id, match_, highlight));
@@ -333,7 +337,7 @@ impl<'c, 'f, 'd> QueryBuilder<'c, 'f, 'd> {
 
     pub fn query(
         self,
-        reader: &heed::RoTxn,
+        reader: &heed::RoTxn<MainT>,
         query: &str,
         range: Range<usize>,
     ) -> MResult<Vec<Document>> {
@@ -371,7 +375,7 @@ impl<'c, 'f, 'd> QueryBuilder<'c, 'f, 'd> {
 }
 
 fn raw_query<'c, FI>(
-    reader: &heed::RoTxn,
+    reader: &heed::RoTxn<MainT>,
 
     query: &str,
     range: Range<usize>,
@@ -507,7 +511,7 @@ where
 }
 
 fn raw_query_with_distinct<'c, FI, FD>(
-    reader: &heed::RoTxn,
+    reader: &heed::RoTxn<MainT>,
 
     query: &str,
     range: Range<usize>,
@@ -692,7 +696,7 @@ mod tests {
     use std::iter::FromIterator;
 
     use fst::{IntoStreamer, Set};
-    use meilidb_schema::SchemaAttr;
+    use meilisearch_schema::SchemaAttr;
     use sdset::SetBuf;
     use tempfile::TempDir;
 
@@ -762,8 +766,8 @@ mod tests {
         }
 
         pub fn add_synonym(&mut self, word: &str, new: SetBuf<&str>) {
-            let env = &self.database.env;
-            let mut writer = env.write_txn().unwrap();
+            let db = &self.database;
+            let mut writer = db.main_write_txn().unwrap();
 
             let word = word.to_lowercase();
 
@@ -806,8 +810,8 @@ mod tests {
             let database = Database::open_or_create(&tempdir).unwrap();
             let index = database.create_index("default").unwrap();
 
-            let env = &database.env;
-            let mut writer = env.write_txn().unwrap();
+            let db = &database;
+            let mut writer = db.main_write_txn().unwrap();
 
             let mut words_fst = BTreeSet::new();
             let mut postings_lists = HashMap::new();
@@ -869,8 +873,8 @@ mod tests {
             ("apple", &[doc_char_index(0, 2, 2)][..]),
         ]);
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "iphone from apple", 0..20).unwrap();
@@ -892,8 +896,8 @@ mod tests {
 
         store.add_synonym("bonjour", SetBuf::from_dirty(vec!["hello"]));
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "hello", 0..20).unwrap();
@@ -925,8 +929,8 @@ mod tests {
         store.add_synonym("bonjour", SetBuf::from_dirty(vec!["hello"]));
         store.add_synonym("salut", SetBuf::from_dirty(vec!["hello"]));
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "sal", 0..20).unwrap();
@@ -969,8 +973,8 @@ mod tests {
 
         store.add_synonym("salutation", SetBuf::from_dirty(vec!["hello"]));
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "salutution", 0..20).unwrap();
@@ -1007,8 +1011,8 @@ mod tests {
         store.add_synonym("bonjour", SetBuf::from_dirty(vec!["hello", "salut"]));
         store.add_synonym("salut", SetBuf::from_dirty(vec!["hello", "bonjour"]));
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "hello", 0..20).unwrap();
@@ -1095,8 +1099,8 @@ mod tests {
             SetBuf::from_dirty(vec!["NY", "new york", "new york city"]),
         );
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "NY subway", 0..20).unwrap();
@@ -1165,8 +1169,8 @@ mod tests {
 
         store.add_synonym("NY", SetBuf::from_dirty(vec!["york new"]));
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "NY", 0..20).unwrap();
@@ -1223,8 +1227,8 @@ mod tests {
 
         store.add_synonym("new york", SetBuf::from_dirty(vec!["NY"]));
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "NY subway", 0..20).unwrap();
@@ -1288,8 +1292,8 @@ mod tests {
             SetBuf::from_dirty(vec!["NY", "new york", "new york city"]),
         );
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "NY subway", 0..20).unwrap();
@@ -1369,8 +1373,8 @@ mod tests {
         );
         store.add_synonym("subway", SetBuf::from_dirty(vec!["underground train"]));
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "NY subway broken", 0..20).unwrap();
@@ -1456,8 +1460,8 @@ mod tests {
         );
         store.add_synonym("underground train", SetBuf::from_dirty(vec!["subway"]));
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder
@@ -1556,8 +1560,8 @@ mod tests {
         store.add_synonym("new york", SetBuf::from_dirty(vec!["new york city"]));
         store.add_synonym("new york city", SetBuf::from_dirty(vec!["new york"]));
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "new york big ", 0..20).unwrap();
@@ -1593,8 +1597,8 @@ mod tests {
 
         store.add_synonym("NY", SetBuf::from_dirty(vec!["new york city story"]));
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "NY subway ", 0..20).unwrap();
@@ -1643,8 +1647,8 @@ mod tests {
         store.add_synonym("new york city", SetBuf::from_dirty(vec!["NYC"]));
         store.add_synonym("subway", SetBuf::from_dirty(vec!["underground train"]));
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder
@@ -1669,15 +1673,15 @@ mod tests {
     #[test]
     fn deunicoded_synonyms() {
         let mut store = TempDatabase::from_iter(vec![
-            ("telephone", &[doc_index(0, 0)][..]), // meilidb indexes the unidecoded
+            ("telephone", &[doc_index(0, 0)][..]), // meilisearch indexes the unidecoded
             ("téléphone", &[doc_index(0, 0)][..]), // and the original words on the same DocIndex
             ("iphone", &[doc_index(1, 0)][..]),
         ]);
 
         store.add_synonym("téléphone", SetBuf::from_dirty(vec!["iphone"]));
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "telephone", 0..20).unwrap();
@@ -1738,8 +1742,8 @@ mod tests {
             ("case", &[doc_index(0, 1)][..]),
         ]);
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "i phone case", 0..20).unwrap();
@@ -1766,8 +1770,8 @@ mod tests {
             ("engine", &[doc_index(1, 2)][..]),
         ]);
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "searchengine", 0..20).unwrap();
@@ -1798,8 +1802,8 @@ mod tests {
             ("engine", &[doc_index(1, 3)][..]),
         ]);
 
-        let env = &store.database.env;
-        let reader = env.read_txn().unwrap();
+        let db = &store.database;
+        let reader = db.main_read_txn().unwrap();
 
         let builder = store.query_builder();
         let results = builder.query(&reader, "searchengine", 0..20).unwrap();

meilisearch-core/src/ranked_map.rs

@@ -1,7 +1,7 @@
 use std::io::{Read, Write};
 
 use hashbrown::HashMap;
-use meilidb_schema::SchemaAttr;
+use meilisearch_schema::SchemaAttr;
 use serde::{Deserialize, Serialize};
 
 use crate::{DocumentId, Number};

meilisearch-core/src/raw_document.rs

@@ -1,7 +1,7 @@
 use std::fmt;
 use std::sync::Arc;
 
-use meilidb_schema::SchemaAttr;
+use meilisearch_schema::SchemaAttr;
 use sdset::SetBuf;
 use slice_group_by::GroupBy;
 

meilisearch-core/src/raw_indexer.rs

@@ -3,10 +3,12 @@ use std::convert::TryFrom;
 
 use crate::{DocIndex, DocumentId};
 use deunicode::deunicode_with_tofu;
-use meilidb_schema::SchemaAttr;
-use meilidb_tokenizer::{is_cjk, SeqTokenizer, Token, Tokenizer};
+use meilisearch_schema::SchemaAttr;
+use meilisearch_tokenizer::{is_cjk, SeqTokenizer, Token, Tokenizer};
 use sdset::SetBuf;
 
+const WORD_LENGTH_LIMIT: usize = 80;
+
 type Word = Vec<u8>; // TODO make it be a SmallVec
 
 pub struct RawIndexer {
@@ -37,54 +39,8 @@ impl RawIndexer {
 
     pub fn index_text(&mut self, id: DocumentId, attr: SchemaAttr, text: &str) -> usize {
         let mut number_of_words = 0;
-        let lowercase_text = text.to_lowercase();
-        let deunicoded = deunicode_with_tofu(&lowercase_text, "");
 
-        // TODO compute the deunicoded version after the cjk check
-        let next = if !lowercase_text.contains(is_cjk) && lowercase_text != deunicoded {
-            Some(deunicoded)
-        } else {
-            None
-        };
-        let iter = Some(lowercase_text).into_iter().chain(next);
-
-        for text in iter {
-            // we must not count 2 times the same words
-            number_of_words = 0;
-
-            for token in Tokenizer::new(&text) {
-                let must_continue = index_token(
-                    token,
-                    id,
-                    attr,
-                    self.word_limit,
-                    &self.stop_words,
-                    &mut self.words_doc_indexes,
-                    &mut self.docs_words,
-                );
-
-                if !must_continue {
-                    break;
-                }
-
-                number_of_words += 1;
-            }
-        }
-
-        number_of_words
-    }
-
-    pub fn index_text_seq<'a, I, IT>(&mut self, id: DocumentId, attr: SchemaAttr, iter: I)
-    where
-        I: IntoIterator<Item = &'a str, IntoIter = IT>,
-        IT: Iterator<Item = &'a str> + Clone,
-    {
-        // TODO serialize this to one call to the SeqTokenizer loop
-
-        let lowercased: Vec<_> = iter.into_iter().map(str::to_lowercase).collect();
-        let iter = lowercased.iter().map(|t| t.as_str());
-
-        for token in SeqTokenizer::new(iter) {
+        for token in Tokenizer::new(text) {
             let must_continue = index_token(
                 token,
                 id,
@@ -95,27 +51,21 @@ impl RawIndexer {
                 &mut self.docs_words,
             );
 
+            number_of_words += 1;
+
             if !must_continue {
                 break;
             }
         }
 
-        let deunicoded: Vec<_> = lowercased
-            .into_iter()
-            .map(|lowercase_text| {
-                if lowercase_text.contains(is_cjk) {
-                    return lowercase_text;
-                }
-                let deunicoded = deunicode_with_tofu(&lowercase_text, "");
-                if lowercase_text != deunicoded {
-                    deunicoded
-                } else {
-                    lowercase_text
-                }
-            })
-            .collect();
-        let iter = deunicoded.iter().map(|t| t.as_str());
+        number_of_words
+    }
 
+    pub fn index_text_seq<'a, I>(&mut self, id: DocumentId, attr: SchemaAttr, iter: I)
+    where
+        I: IntoIterator<Item = &'a str>,
+    {
+        let iter = iter.into_iter();
         for token in SeqTokenizer::new(iter) {
             let must_continue = index_token(
                 token,
@@ -170,15 +120,38 @@ fn index_token(
         return false;
     }
 
+    let lower = token.word.to_lowercase();
+    let token = Token {
+        word: &lower,
+        ..token
+    };
+
     if !stop_words.contains(&token.word) {
         match token_to_docindex(id, attr, token) {
             Some(docindex) => {
                 let word = Vec::from(token.word);
-                words_doc_indexes
-                    .entry(word.clone())
-                    .or_insert_with(Vec::new)
-                    .push(docindex);
-                docs_words.entry(id).or_insert_with(Vec::new).push(word);
+
+                if word.len() <= WORD_LENGTH_LIMIT {
+                    words_doc_indexes
+                        .entry(word.clone())
+                        .or_insert_with(Vec::new)
+                        .push(docindex);
+                    docs_words.entry(id).or_insert_with(Vec::new).push(word);
+
+                    if !lower.contains(is_cjk) {
+                        let unidecoded = deunicode_with_tofu(&lower, "");
+                        if unidecoded != lower && !unidecoded.is_empty() {
+                            let word = Vec::from(unidecoded);
+                            if word.len() <= WORD_LENGTH_LIMIT {
+                                words_doc_indexes
+                                    .entry(word.clone())
+                                    .or_insert_with(Vec::new)
+                                    .push(docindex);
+                                docs_words.entry(id).or_insert_with(Vec::new).push(word);
+                            }
+                        }
+                    }
+                }
             }
             None => return false,
         }
@@ -224,10 +197,8 @@ mod tests {
         assert!(words_doc_indexes.get(&b"aspirateur"[..]).is_some());
         assert!(words_doc_indexes.get(&b"ai"[..]).is_some());
         assert!(words_doc_indexes.get(&b"eteindre"[..]).is_some());
-
-        // with the ugly apostrophe...
         assert!(words_doc_indexes
-            .get(&"l’éteindre".to_owned().into_bytes())
+            .get(&"éteindre".to_owned().into_bytes())
             .is_some());
     }
 
@@ -248,10 +219,8 @@ mod tests {
         assert!(words_doc_indexes.get(&b"aspirateur"[..]).is_some());
         assert!(words_doc_indexes.get(&b"ai"[..]).is_some());
         assert!(words_doc_indexes.get(&b"eteindre"[..]).is_some());
-
-        // with the ugly apostrophe...
         assert!(words_doc_indexes
-            .get(&"l’éteindre".to_owned().into_bytes())
+            .get(&"éteindre".to_owned().into_bytes())
             .is_some());
     }
 
@@ -277,10 +246,26 @@ mod tests {
         assert!(words_doc_indexes.get(&b"ai"[..]).is_none());
         assert!(words_doc_indexes.get(&b"de"[..]).is_none());
         assert!(words_doc_indexes.get(&b"eteindre"[..]).is_some());
-
-        // with the ugly apostrophe...
         assert!(words_doc_indexes
-            .get(&"l’éteindre".to_owned().into_bytes())
+            .get(&"éteindre".to_owned().into_bytes())
+            .is_some());
+    }
+
+    #[test]
+    fn no_empty_unidecode() {
+        let mut indexer = RawIndexer::new(fst::Set::default());
+
+        let docid = DocumentId(0);
+        let attr = SchemaAttr(0);
+        let text = "🇯🇵";
+        indexer.index_text(docid, attr, text);
+
+        let Indexed {
+            words_doc_indexes, ..
+        } = indexer.build();
+
+        assert!(words_doc_indexes
+            .get(&"🇯🇵".to_owned().into_bytes())
             .is_some());
     }
 }

meilisearch-core/src/serde/deserializer.rs

@@ -2,12 +2,13 @@ use std::collections::HashSet;
 use std::io::Cursor;
 use std::{error::Error, fmt};
 
-use meilidb_schema::{Schema, SchemaAttr};
+use meilisearch_schema::{Schema, SchemaAttr};
 use serde::{de, forward_to_deserialize_any};
 use serde_json::de::IoRead as SerdeJsonIoRead;
 use serde_json::Deserializer as SerdeJsonDeserializer;
 use serde_json::Error as SerdeJsonError;
 
+use crate::database::MainT;
 use crate::store::DocumentsFields;
 use crate::DocumentId;
 
@@ -50,7 +51,7 @@ impl From<heed::Error> for DeserializerError {
 
 pub struct Deserializer<'a> {
     pub document_id: DocumentId,
-    pub reader: &'a heed::RoTxn,
+    pub reader: &'a heed::RoTxn<MainT>,
     pub documents_fields: DocumentsFields,
     pub schema: &'a Schema,
     pub attributes: Option<&'a HashSet<SchemaAttr>>,
@@ -63,13 +64,14 @@ impl<'de, 'a, 'b> de::Deserializer<'de> for &'b mut Deserializer<'a> {
     where
         V: de::Visitor<'de>,
     {
-        self.deserialize_map(visitor)
+        self.deserialize_option(visitor)
     }
 
-    forward_to_deserialize_any! {
-        bool i8 i16 i32 i64 i128 u8 u16 u32 u64 u128 f32 f64 char str string
-        bytes byte_buf option unit unit_struct newtype_struct seq tuple
-        tuple_struct struct enum identifier ignored_any
+    fn deserialize_option<V>(self, visitor: V) -> Result<V::Value, Self::Error>
+    where
+        V: de::Visitor<'de>,
+    {
+        self.deserialize_map(visitor)
     }
 
     fn deserialize_map<V>(self, visitor: V) -> Result<V::Value, Self::Error>
@@ -104,16 +106,29 @@ impl<'de, 'a, 'b> de::Deserializer<'de> for &'b mut Deserializer<'a> {
                 }
             });
 
-        let map_deserializer = de::value::MapDeserializer::new(iter);
-        let result = visitor
-            .visit_map(map_deserializer)
-            .map_err(DeserializerError::from);
+        let mut iter = iter.peekable();
+
+        let result = match iter.peek() {
+            Some(_) => {
+                let map_deserializer = de::value::MapDeserializer::new(iter);
+                visitor
+                    .visit_some(map_deserializer)
+                    .map_err(DeserializerError::from)
+            }
+            None => visitor.visit_none(),
+        };
 
         match error.take() {
             Some(error) => Err(error.into()),
             None => result,
         }
     }
+
+    forward_to_deserialize_any! {
+        bool i8 i16 i32 i64 i128 u8 u16 u32 u64 u128 f32 f64 char str string
+        bytes byte_buf unit unit_struct newtype_struct seq tuple
+        tuple_struct struct enum identifier ignored_any
+    }
 }
 
 struct Value(SerdeJsonDeserializer<SerdeJsonIoRead<Cursor<Vec<u8>>>>);

meilisearch-core/src/serde/indexer.rs

@@ -1,4 +1,4 @@
-use meilidb_schema::SchemaAttr;
+use meilisearch_schema::SchemaAttr;
 use serde::ser;
 use serde::Serialize;
 

meilisearch-core/src/serde/serializer.rs

@@ -1,14 +1,15 @@
-use meilidb_schema::{Schema, SchemaAttr, SchemaProps};
+use meilisearch_schema::{Schema, SchemaAttr, SchemaProps};
 use serde::ser;
 
+use crate::database::MainT;
 use crate::raw_indexer::RawIndexer;
 use crate::store::{DocumentsFields, DocumentsFieldsCounts};
 use crate::{DocumentId, RankedMap};
 
 use super::{ConvertToNumber, ConvertToString, Indexer, SerializerError};
 
-pub struct Serializer<'a> {
-    pub txn: &'a mut heed::RwTxn,
+pub struct Serializer<'a, 'b> {
+    pub txn: &'a mut heed::RwTxn<'b, MainT>,
     pub schema: &'a Schema,
     pub document_store: DocumentsFields,
     pub document_fields_counts: DocumentsFieldsCounts,
@@ -17,15 +18,15 @@ pub struct Serializer<'a> {
     pub document_id: DocumentId,
 }
 
-impl<'a> ser::Serializer for Serializer<'a> {
+impl<'a, 'b> ser::Serializer for Serializer<'a, 'b> {
     type Ok = ();
     type Error = SerializerError;
     type SerializeSeq = ser::Impossible<Self::Ok, Self::Error>;
     type SerializeTuple = ser::Impossible<Self::Ok, Self::Error>;
     type SerializeTupleStruct = ser::Impossible<Self::Ok, Self::Error>;
     type SerializeTupleVariant = ser::Impossible<Self::Ok, Self::Error>;
-    type SerializeMap = MapSerializer<'a>;
-    type SerializeStruct = StructSerializer<'a>;
+    type SerializeMap = MapSerializer<'a, 'b>;
+    type SerializeStruct = StructSerializer<'a, 'b>;
     type SerializeStructVariant = ser::Impossible<Self::Ok, Self::Error>;
 
     forward_to_unserializable_type! {
@@ -190,8 +191,8 @@ impl<'a> ser::Serializer for Serializer<'a> {
     }
 }
 
-pub struct MapSerializer<'a> {
-    txn: &'a mut heed::RwTxn,
+pub struct MapSerializer<'a, 'b> {
+    txn: &'a mut heed::RwTxn<'b, MainT>,
     schema: &'a Schema,
     document_id: DocumentId,
     document_store: DocumentsFields,
@@ -201,7 +202,7 @@ pub struct MapSerializer<'a> {
     current_key_name: Option<String>,
 }
 
-impl<'a> ser::SerializeMap for MapSerializer<'a> {
+impl<'a, 'b> ser::SerializeMap for MapSerializer<'a, 'b> {
     type Ok = ();
     type Error = SerializerError;
 
@@ -253,8 +254,8 @@ impl<'a> ser::SerializeMap for MapSerializer<'a> {
     }
 }
 
-pub struct StructSerializer<'a> {
-    txn: &'a mut heed::RwTxn,
+pub struct StructSerializer<'a, 'b> {
+    txn: &'a mut heed::RwTxn<'b, MainT>,
     schema: &'a Schema,
     document_id: DocumentId,
     document_store: DocumentsFields,
@@ -263,7 +264,7 @@ pub struct StructSerializer<'a> {
     ranked_map: &'a mut RankedMap,
 }
 
-impl<'a> ser::SerializeStruct for StructSerializer<'a> {
+impl<'a, 'b> ser::SerializeStruct for StructSerializer<'a, 'b> {
     type Ok = ();
     type Error = SerializerError;
 
@@ -297,7 +298,7 @@ impl<'a> ser::SerializeStruct for StructSerializer<'a> {
 }
 
 pub fn serialize_value<T: ?Sized>(
-    txn: &mut heed::RwTxn,
+    txn: &mut heed::RwTxn<MainT>,
     attribute: SchemaAttr,
     props: SchemaProps,
     document_id: DocumentId,

meilisearch-core/src/store/docs_words.rs

@@ -1,4 +1,5 @@
 use super::BEU64;
+use crate::database::MainT;
 use crate::DocumentId;
 use heed::types::{ByteSlice, OwnedType};
 use heed::Result as ZResult;
@@ -12,7 +13,7 @@ pub struct DocsWords {
 impl DocsWords {
     pub fn put_doc_words(
         self,
-        writer: &mut heed::RwTxn,
+        writer: &mut heed::RwTxn<MainT>,
         document_id: DocumentId,
         words: &fst::Set,
     ) -> ZResult<()> {
@@ -21,25 +22,25 @@ impl DocsWords {
         self.docs_words.put(writer, &document_id, bytes)
     }
 
-    pub fn del_doc_words(self, writer: &mut heed::RwTxn, document_id: DocumentId) -> ZResult<bool> {
+    pub fn del_doc_words(self, writer: &mut heed::RwTxn<MainT>, document_id: DocumentId) -> ZResult<bool> {
         let document_id = BEU64::new(document_id.0);
         self.docs_words.delete(writer, &document_id)
     }
 
-    pub fn clear(self, writer: &mut heed::RwTxn) -> ZResult<()> {
+    pub fn clear(self, writer: &mut heed::RwTxn<MainT>) -> ZResult<()> {
         self.docs_words.clear(writer)
     }
 
     pub fn doc_words(
         self,
-        reader: &heed::RoTxn,
+        reader: &heed::RoTxn<MainT>,
         document_id: DocumentId,
     ) -> ZResult<Option<fst::Set>> {
         let document_id = BEU64::new(document_id.0);
         match self.docs_words.get(reader, &document_id)? {
             Some(bytes) => {
                 let len = bytes.len();
-                let bytes = Arc::from(bytes);
+                let bytes = Arc::new(bytes.to_owned());
                 let fst = fst::raw::Fst::from_shared_bytes(bytes, 0, len).unwrap();
                 Ok(Some(fst::Set::from(fst)))
             }

meilisearch-core/src/store/documents_fields.rs

@@ -1,6 +1,7 @@
 use heed::types::{ByteSlice, OwnedType};
+use crate::database::MainT;
 use heed::Result as ZResult;
-use meilidb_schema::SchemaAttr;
+use meilisearch_schema::SchemaAttr;
 
 use super::DocumentAttrKey;
 use crate::DocumentId;
@@ -13,7 +14,7 @@ pub struct DocumentsFields {
 impl DocumentsFields {
     pub fn put_document_field(
         self,
-        writer: &mut heed::RwTxn,
+        writer: &mut heed::RwTxn<MainT>,
         document_id: DocumentId,
         attribute: SchemaAttr,
         value: &[u8],
@@ -24,21 +25,21 @@ impl DocumentsFields {
 
     pub fn del_all_document_fields(
         self,
-        writer: &mut heed::RwTxn,
+        writer: &mut heed::RwTxn<MainT>,
         document_id: DocumentId,
     ) -> ZResult<usize> {
         let start = DocumentAttrKey::new(document_id, SchemaAttr::min());
         let end = DocumentAttrKey::new(document_id, SchemaAttr::max());
-        self.documents_fields.delete_range(writer, start..=end)
+        self.documents_fields.delete_range(writer, &(start..=end))
     }
 
-    pub fn clear(self, writer: &mut heed::RwTxn) -> ZResult<()> {
+    pub fn clear(self, writer: &mut heed::RwTxn<MainT>) -> ZResult<()> {
         self.documents_fields.clear(writer)
     }
 
     pub fn document_attribute<'txn>(
         self,
-        reader: &'txn heed::RoTxn,
+        reader: &'txn heed::RoTxn<MainT>,
         document_id: DocumentId,
         attribute: SchemaAttr,
     ) -> ZResult<Option<&'txn [u8]>> {
@@ -48,12 +49,12 @@ impl DocumentsFields {
 
     pub fn document_fields<'txn>(
         self,
-        reader: &'txn heed::RoTxn,
+        reader: &'txn heed::RoTxn<MainT>,
         document_id: DocumentId,
     ) -> ZResult<DocumentFieldsIter<'txn>> {
         let start = DocumentAttrKey::new(document_id, SchemaAttr::min());
         let end = DocumentAttrKey::new(document_id, SchemaAttr::max());
-        let iter = self.documents_fields.range(reader, start..=end)?;
+        let iter = self.documents_fields.range(reader, &(start..=end))?;
         Ok(DocumentFieldsIter { iter })
     }
 }

meilisearch-core/src/store/documents_fields_counts.rs

@@ -1,8 +1,9 @@
 use super::DocumentAttrKey;
+use crate::database::MainT;
 use crate::DocumentId;
 use heed::types::OwnedType;
 use heed::Result as ZResult;
-use meilidb_schema::SchemaAttr;
+use meilisearch_schema::SchemaAttr;
 
 #[derive(Copy, Clone)]
 pub struct DocumentsFieldsCounts {
@@ -12,7 +13,7 @@ pub struct DocumentsFieldsCounts {
 impl DocumentsFieldsCounts {
     pub fn put_document_field_count(
         self,
-        writer: &mut heed::RwTxn,
+        writer: &mut heed::RwTxn<MainT>,
         document_id: DocumentId,
         attribute: SchemaAttr,
         value: u64,
@@ -23,22 +24,22 @@ impl DocumentsFieldsCounts {
 
     pub fn del_all_document_fields_counts(
         self,
-        writer: &mut heed::RwTxn,
+        writer: &mut heed::RwTxn<MainT>,
         document_id: DocumentId,
     ) -> ZResult<usize> {
         let start = DocumentAttrKey::new(document_id, SchemaAttr::min());
         let end = DocumentAttrKey::new(document_id, SchemaAttr::max());
         self.documents_fields_counts
-            .delete_range(writer, start..=end)
+            .delete_range(writer, &(start..=end))
     }
 
-    pub fn clear(self, writer: &mut heed::RwTxn) -> ZResult<()> {
+    pub fn clear(self, writer: &mut heed::RwTxn<MainT>) -> ZResult<()> {
         self.documents_fields_counts.clear(writer)
     }
 
     pub fn document_field_count(
         self,
-        reader: &heed::RoTxn,
+        reader: &heed::RoTxn<MainT>,
         document_id: DocumentId,
         attribute: SchemaAttr,
     ) -> ZResult<Option<u64>> {
@@ -51,16 +52,16 @@ impl DocumentsFieldsCounts {
 
     pub fn document_fields_counts<'txn>(
         self,
-        reader: &'txn heed::RoTxn,
+        reader: &'txn heed::RoTxn<MainT>,
         document_id: DocumentId,
     ) -> ZResult<DocumentFieldsCountsIter<'txn>> {
         let start = DocumentAttrKey::new(document_id, SchemaAttr::min());
         let end = DocumentAttrKey::new(document_id, SchemaAttr::max());
-        let iter = self.documents_fields_counts.range(reader, start..=end)?;
+        let iter = self.documents_fields_counts.range(reader, &(start..=end))?;
         Ok(DocumentFieldsCountsIter { iter })
     }
 
-    pub fn documents_ids<'txn>(self, reader: &'txn heed::RoTxn) -> ZResult<DocumentsIdsIter<'txn>> {
+    pub fn documents_ids<'txn>(self, reader: &'txn heed::RoTxn<MainT>) -> ZResult<DocumentsIdsIter<'txn>> {
         let iter = self.documents_fields_counts.iter(reader)?;
         Ok(DocumentsIdsIter {
             last_seen_id: None,
@@ -70,7 +71,7 @@ impl DocumentsFieldsCounts {
 
     pub fn all_documents_fields_counts<'txn>(
         self,
-        reader: &'txn heed::RoTxn,
+        reader: &'txn heed::RoTxn<MainT>,
     ) -> ZResult<AllDocumentsFieldsCountsIter<'txn>> {
         let iter = self.documents_fields_counts.iter(reader)?;
         Ok(AllDocumentsFieldsCountsIter { iter })

meilisearch-core/src/store/main.rs

@@ -0,0 +1,184 @@
+use crate::database::MainT;
+use crate::RankedMap;
+use chrono::{DateTime, Utc};
+use heed::types::{ByteSlice, OwnedType, SerdeBincode, Str};
+use heed::Result as ZResult;
+use meilisearch_schema::Schema;
+use std::collections::HashMap;
+use std::sync::Arc;
+
+const CREATED_AT_KEY: &str = "created-at";
+const CUSTOMS_KEY: &str = "customs-key";
+const FIELDS_FREQUENCY_KEY: &str = "fields-frequency";
+const NAME_KEY: &str = "name";
+const NUMBER_OF_DOCUMENTS_KEY: &str = "number-of-documents";
+const RANKED_MAP_KEY: &str = "ranked-map";
+const SCHEMA_KEY: &str = "schema";
+const STOP_WORDS_KEY: &str = "stop-words";
+const SYNONYMS_KEY: &str = "synonyms";
+const UPDATED_AT_KEY: &str = "updated-at";
+const WORDS_KEY: &str = "words";
+
+pub type FreqsMap = HashMap<String, usize>;
+type SerdeFreqsMap = SerdeBincode<FreqsMap>;
+type SerdeDatetime = SerdeBincode<DateTime<Utc>>;
+
+#[derive(Copy, Clone)]
+pub struct Main {
+    pub(crate) main: heed::PolyDatabase,
+}
+
+impl Main {
+    pub fn clear(self, writer: &mut heed::RwTxn<MainT>) -> ZResult<()> {
+        self.main.clear(writer)
+    }
+
+    pub fn put_name(self, writer: &mut heed::RwTxn<MainT>, name: &str) -> ZResult<()> {
+        self.main.put::<_, Str, Str>(writer, NAME_KEY, name)
+    }
+
+    pub fn name(self, reader: &heed::RoTxn<MainT>) -> ZResult<Option<String>> {
+        Ok(self
+            .main
+            .get::<_, Str, Str>(reader, NAME_KEY)?
+            .map(|name| name.to_owned()))
+    }
+
+    pub fn put_created_at(self, writer: &mut heed::RwTxn<MainT>) -> ZResult<()> {
+        self.main
+            .put::<_, Str, SerdeDatetime>(writer, CREATED_AT_KEY, &Utc::now())
+    }
+
+    pub fn created_at(self, reader: &heed::RoTxn<MainT>) -> ZResult<Option<DateTime<Utc>>> {
+        self.main.get::<_, Str, SerdeDatetime>(reader, CREATED_AT_KEY)
+    }
+
+    pub fn put_updated_at(self, writer: &mut heed::RwTxn<MainT>) -> ZResult<()> {
+        self.main
+            .put::<_, Str, SerdeDatetime>(writer, UPDATED_AT_KEY, &Utc::now())
+    }
+
+    pub fn updated_at(self, reader: &heed::RoTxn<MainT>) -> ZResult<Option<DateTime<Utc>>> {
+        self.main.get::<_, Str, SerdeDatetime>(reader, UPDATED_AT_KEY)
+    }
+
+    pub fn put_words_fst(self, writer: &mut heed::RwTxn<MainT>, fst: &fst::Set) -> ZResult<()> {
+        let bytes = fst.as_fst().as_bytes();
+        self.main.put::<_, Str, ByteSlice>(writer, WORDS_KEY, bytes)
+    }
+
+    pub fn words_fst(self, reader: &heed::RoTxn<MainT>) -> ZResult<Option<fst::Set>> {
+        match self.main.get::<_, Str, ByteSlice>(reader, WORDS_KEY)? {
+            Some(bytes) => {
+                let len = bytes.len();
+                let bytes = Arc::new(bytes.to_owned());
+                let fst = fst::raw::Fst::from_shared_bytes(bytes, 0, len).unwrap();
+                Ok(Some(fst::Set::from(fst)))
+            }
+            None => Ok(None),
+        }
+    }
+
+    pub fn put_schema(self, writer: &mut heed::RwTxn<MainT>, schema: &Schema) -> ZResult<()> {
+        self.main
+            .put::<_, Str, SerdeBincode<Schema>>(writer, SCHEMA_KEY, schema)
+    }
+
+    pub fn schema(self, reader: &heed::RoTxn<MainT>) -> ZResult<Option<Schema>> {
+        self.main
+            .get::<_, Str, SerdeBincode<Schema>>(reader, SCHEMA_KEY)
+    }
+
+    pub fn put_ranked_map(self, writer: &mut heed::RwTxn<MainT>, ranked_map: &RankedMap) -> ZResult<()> {
+        self.main
+            .put::<_, Str, SerdeBincode<RankedMap>>(writer, RANKED_MAP_KEY, &ranked_map)
+    }
+
+    pub fn ranked_map(self, reader: &heed::RoTxn<MainT>) -> ZResult<Option<RankedMap>> {
+        self.main
+            .get::<_, Str, SerdeBincode<RankedMap>>(reader, RANKED_MAP_KEY)
+    }
+
+    pub fn put_synonyms_fst(self, writer: &mut heed::RwTxn<MainT>, fst: &fst::Set) -> ZResult<()> {
+        let bytes = fst.as_fst().as_bytes();
+        self.main.put::<_, Str, ByteSlice>(writer, SYNONYMS_KEY, bytes)
+    }
+
+    pub fn synonyms_fst(self, reader: &heed::RoTxn<MainT>) -> ZResult<Option<fst::Set>> {
+        match self.main.get::<_, Str, ByteSlice>(reader, SYNONYMS_KEY)? {
+            Some(bytes) => {
+                let len = bytes.len();
+                let bytes = Arc::new(bytes.to_owned());
+                let fst = fst::raw::Fst::from_shared_bytes(bytes, 0, len).unwrap();
+                Ok(Some(fst::Set::from(fst)))
+            }
+            None => Ok(None),
+        }
+    }
+
+    pub fn put_stop_words_fst(self, writer: &mut heed::RwTxn<MainT>, fst: &fst::Set) -> ZResult<()> {
+        let bytes = fst.as_fst().as_bytes();
+        self.main
+            .put::<_, Str, ByteSlice>(writer, STOP_WORDS_KEY, bytes)
+    }
+
+    pub fn stop_words_fst(self, reader: &heed::RoTxn<MainT>) -> ZResult<Option<fst::Set>> {
+        match self.main.get::<_, Str, ByteSlice>(reader, STOP_WORDS_KEY)? {
+            Some(bytes) => {
+                let len = bytes.len();
+                let bytes = Arc::new(bytes.to_owned());
+                let fst = fst::raw::Fst::from_shared_bytes(bytes, 0, len).unwrap();
+                Ok(Some(fst::Set::from(fst)))
+            }
+            None => Ok(None),
+        }
+    }
+
+    pub fn put_number_of_documents<F>(self, writer: &mut heed::RwTxn<MainT>, f: F) -> ZResult<u64>
+    where
+        F: Fn(u64) -> u64,
+    {
+        let new = self.number_of_documents(&*writer).map(f)?;
+        self.main
+            .put::<_, Str, OwnedType<u64>>(writer, NUMBER_OF_DOCUMENTS_KEY, &new)?;
+        Ok(new)
+    }
+
+    pub fn number_of_documents(self, reader: &heed::RoTxn<MainT>) -> ZResult<u64> {
+        match self
+            .main
+            .get::<_, Str, OwnedType<u64>>(reader, NUMBER_OF_DOCUMENTS_KEY)?
+        {
+            Some(value) => Ok(value),
+            None => Ok(0),
+        }
+    }
+
+    pub fn put_fields_frequency(
+        self,
+        writer: &mut heed::RwTxn<MainT>,
+        fields_frequency: &FreqsMap,
+    ) -> ZResult<()> {
+        self.main
+            .put::<_, Str, SerdeFreqsMap>(writer, FIELDS_FREQUENCY_KEY, fields_frequency)
+    }
+
+    pub fn fields_frequency(&self, reader: &heed::RoTxn<MainT>) -> ZResult<Option<FreqsMap>> {
+        match self
+            .main
+            .get::<_, Str, SerdeFreqsMap>(reader, FIELDS_FREQUENCY_KEY)?
+        {
+            Some(freqs) => Ok(Some(freqs)),
+            None => Ok(None),
+        }
+    }
+
+    pub fn put_customs(self, writer: &mut heed::RwTxn<MainT>, customs: &[u8]) -> ZResult<()> {
+        self.main
+            .put::<_, Str, ByteSlice>(writer, CUSTOMS_KEY, customs)
+    }
+
+    pub fn customs<'txn>(self, reader: &'txn heed::RoTxn<MainT>) -> ZResult<Option<&'txn [u8]>> {
+        self.main.get::<_, Str, ByteSlice>(reader, CUSTOMS_KEY)
+    }
+}

meilisearch-core/src/store/mod.rs

@@ -21,11 +21,13 @@ pub use self::updates_results::UpdatesResults;
 use std::collections::HashSet;
 
 use heed::Result as ZResult;
-use meilidb_schema::{Schema, SchemaAttr};
-use serde::de;
+use meilisearch_schema::{Schema, SchemaAttr};
+use serde::de::{self, Deserialize};
 use zerocopy::{AsBytes, FromBytes};
 
 use crate::criterion::Criteria;
+use crate::database::{UpdateEvent, UpdateEventsEmitter};
+use crate::database::{MainT, UpdateT};
 use crate::serde::Deserializer;
 use crate::{query_builder::QueryBuilder, update, DocumentId, Error, MResult};
 
@@ -91,13 +93,13 @@ pub struct Index {
 
     pub updates: Updates,
     pub updates_results: UpdatesResults,
-    updates_notifier: crossbeam_channel::Sender<()>,
+    pub(crate) updates_notifier: UpdateEventsEmitter,
 }
 
 impl Index {
     pub fn document<T: de::DeserializeOwned>(
         &self,
-        reader: &heed::RoTxn,
+        reader: &heed::RoTxn<MainT>,
         attributes: Option<&HashSet<&str>>,
         document_id: DocumentId,
     ) -> MResult<Option<T>> {
@@ -120,14 +122,12 @@ impl Index {
             attributes: attributes.as_ref(),
         };
 
-        // TODO: currently we return an error if all document fields are missing,
-        //       returning None would have been better
-        Ok(T::deserialize(&mut deserializer).map(Some)?)
+        Ok(Option::<T>::deserialize(&mut deserializer)?)
     }
 
     pub fn document_attribute<T: de::DeserializeOwned>(
         &self,
-        reader: &heed::RoTxn,
+        reader: &heed::RoTxn<MainT>,
         document_id: DocumentId,
         attribute: SchemaAttr,
     ) -> MResult<Option<T>> {
@@ -140,13 +140,13 @@ impl Index {
         }
     }
 
-    pub fn schema_update(&self, writer: &mut heed::RwTxn, schema: Schema) -> MResult<u64> {
-        let _ = self.updates_notifier.send(());
+    pub fn schema_update(&self, writer: &mut heed::RwTxn<UpdateT>, schema: Schema) -> MResult<u64> {
+        let _ = self.updates_notifier.send(UpdateEvent::NewUpdate);
         update::push_schema_update(writer, self.updates, self.updates_results, schema)
     }
 
-    pub fn customs_update(&self, writer: &mut heed::RwTxn, customs: Vec<u8>) -> ZResult<u64> {
-        let _ = self.updates_notifier.send(());
+    pub fn customs_update(&self, writer: &mut heed::RwTxn<UpdateT>, customs: Vec<u8>) -> ZResult<u64> {
+        let _ = self.updates_notifier.send(UpdateEvent::NewUpdate);
         update::push_customs_update(writer, self.updates, self.updates_results, customs)
     }
 
@@ -158,6 +158,14 @@ impl Index {
         )
     }
 
+    pub fn documents_partial_addition<D>(&self) -> update::DocumentsAddition<D> {
+        update::DocumentsAddition::new_partial(
+            self.updates,
+            self.updates_results,
+            self.updates_notifier.clone(),
+        )
+    }
+
     pub fn documents_deletion(&self) -> update::DocumentsDeletion {
         update::DocumentsDeletion::new(
             self.updates,
@@ -166,8 +174,8 @@ impl Index {
         )
     }
 
-    pub fn clear_all(&self, writer: &mut heed::RwTxn) -> MResult<u64> {
-        let _ = self.updates_notifier.send(());
+    pub fn clear_all(&self, writer: &mut heed::RwTxn<UpdateT>) -> MResult<u64> {
+        let _ = self.updates_notifier.send(UpdateEvent::NewUpdate);
         update::push_clear_all(writer, self.updates, self.updates_results)
     }
 
@@ -203,8 +211,8 @@ impl Index {
         )
     }
 
-    pub fn current_update_id(&self, reader: &heed::RoTxn) -> MResult<Option<u64>> {
-        match self.updates.last_update_id(reader)? {
+    pub fn current_update_id(&self, reader: &heed::RoTxn<UpdateT>) -> MResult<Option<u64>> {
+        match self.updates.last_update(reader)? {
             Some((id, _)) => Ok(Some(id)),
             None => Ok(None),
         }
@@ -212,32 +220,34 @@ impl Index {
 
     pub fn update_status(
         &self,
-        reader: &heed::RoTxn,
+        reader: &heed::RoTxn<UpdateT>,
         update_id: u64,
-    ) -> MResult<update::UpdateStatus> {
+    ) -> MResult<Option<update::UpdateStatus>> {
         update::update_status(reader, self.updates, self.updates_results, update_id)
     }
 
-    pub fn all_updates_status(&self, reader: &heed::RoTxn) -> MResult<Vec<update::UpdateStatus>> {
+    pub fn all_updates_status(&self, reader: &heed::RoTxn<UpdateT>) -> MResult<Vec<update::UpdateStatus>> {
         let mut updates = Vec::new();
         let mut last_update_result_id = 0;
 
         // retrieve all updates results
-        if let Some((last_id, _)) = self.updates_results.last_update_id(reader)? {
+        if let Some((last_id, _)) = self.updates_results.last_update(reader)? {
             updates.reserve(last_id as usize);
 
             for id in 0..=last_id {
-                let update = self.update_status(reader, id)?;
-                updates.push(update);
-                last_update_result_id = id;
+                if let Some(update) = self.update_status(reader, id)? {
+                    updates.push(update);
+                    last_update_result_id = id;
+                }
             }
         }
 
         // retrieve all enqueued updates
-        if let Some((last_id, _)) = self.updates.last_update_id(reader)? {
-            for id in last_update_result_id + 1..last_id {
-                let update = self.update_status(reader, id)?;
-                updates.push(update);
+        if let Some((last_id, _)) = self.updates.last_update(reader)? {
+            for id in last_update_result_id + 1..=last_id {
+                if let Some(update) = self.update_status(reader, id)? {
+                    updates.push(update);
+                }
             }
         }
 
@@ -269,8 +279,9 @@ impl Index {
 
 pub fn create(
     env: &heed::Env,
+    update_env: &heed::Env,
     name: &str,
-    updates_notifier: crossbeam_channel::Sender<()>,
+    updates_notifier: UpdateEventsEmitter,
 ) -> MResult<Index> {
     // create all the store names
     let main_name = main_name(name);
@@ -283,14 +294,14 @@ pub fn create(
     let updates_results_name = updates_results_name(name);
 
     // open all the stores
-    let main = env.create_dyn_database(Some(&main_name))?;
+    let main = env.create_poly_database(Some(&main_name))?;
     let postings_lists = env.create_database(Some(&postings_lists_name))?;
     let documents_fields = env.create_database(Some(&documents_fields_name))?;
     let documents_fields_counts = env.create_database(Some(&documents_fields_counts_name))?;
     let synonyms = env.create_database(Some(&synonyms_name))?;
     let docs_words = env.create_database(Some(&docs_words_name))?;
-    let updates = env.create_database(Some(&updates_name))?;
-    let updates_results = env.create_database(Some(&updates_results_name))?;
+    let updates = update_env.create_database(Some(&updates_name))?;
+    let updates_results = update_env.create_database(Some(&updates_results_name))?;
 
     Ok(Index {
         main: Main { main },
@@ -309,8 +320,9 @@ pub fn create(
 
 pub fn open(
     env: &heed::Env,
+    update_env: &heed::Env,
     name: &str,
-    updates_notifier: crossbeam_channel::Sender<()>,
+    updates_notifier: UpdateEventsEmitter,
 ) -> MResult<Option<Index>> {
     // create all the store names
     let main_name = main_name(name);
@@ -323,7 +335,7 @@ pub fn open(
     let updates_results_name = updates_results_name(name);
 
     // open all the stores
-    let main = match env.open_dyn_database(Some(&main_name))? {
+    let main = match env.open_poly_database(Some(&main_name))? {
         Some(main) => main,
         None => return Ok(None),
     };
@@ -347,11 +359,11 @@ pub fn open(
         Some(docs_words) => docs_words,
         None => return Ok(None),
     };
-    let updates = match env.open_database(Some(&updates_name))? {
+    let updates = match update_env.open_database(Some(&updates_name))? {
         Some(updates) => updates,
         None => return Ok(None),
     };
-    let updates_results = match env.open_database(Some(&updates_results_name))? {
+    let updates_results = match update_env.open_database(Some(&updates_results_name))? {
         Some(updates_results) => updates_results,
         None => return Ok(None),
     };
@@ -370,3 +382,20 @@ pub fn open(
         updates_notifier,
     }))
 }
+
+pub fn clear(
+    writer: &mut heed::RwTxn<MainT>,
+    update_writer: &mut heed::RwTxn<UpdateT>,
+    index: &Index,
+) -> MResult<()> {
+    // clear all the stores
+    index.main.clear(writer)?;
+    index.postings_lists.clear(writer)?;
+    index.documents_fields.clear(writer)?;
+    index.documents_fields_counts.clear(writer)?;
+    index.synonyms.clear(writer)?;
+    index.docs_words.clear(writer)?;
+    index.updates.clear(update_writer)?;
+    index.updates_results.clear(update_writer)?;
+    Ok(())
+}

meilisearch-core/src/store/postings_lists.rs

@@ -1,4 +1,5 @@
 use crate::DocIndex;
+use crate::database::MainT;
 use heed::types::{ByteSlice, CowSlice};
 use heed::Result as ZResult;
 use sdset::{Set, SetBuf};
@@ -12,24 +13,24 @@ pub struct PostingsLists {
 impl PostingsLists {
     pub fn put_postings_list(
         self,
-        writer: &mut heed::RwTxn,
+        writer: &mut heed::RwTxn<MainT>,
         word: &[u8],
         words_indexes: &Set<DocIndex>,
     ) -> ZResult<()> {
         self.postings_lists.put(writer, word, words_indexes)
     }
 
-    pub fn del_postings_list(self, writer: &mut heed::RwTxn, word: &[u8]) -> ZResult<bool> {
+    pub fn del_postings_list(self, writer: &mut heed::RwTxn<MainT>, word: &[u8]) -> ZResult<bool> {
         self.postings_lists.delete(writer, word)
     }
 
-    pub fn clear(self, writer: &mut heed::RwTxn) -> ZResult<()> {
+    pub fn clear(self, writer: &mut heed::RwTxn<MainT>) -> ZResult<()> {
         self.postings_lists.clear(writer)
     }
 
     pub fn postings_list<'txn>(
         self,
-        reader: &'txn heed::RoTxn,
+        reader: &'txn heed::RoTxn<MainT>,
         word: &[u8],
     ) -> ZResult<Option<Cow<'txn, Set<DocIndex>>>> {
         match self.postings_lists.get(reader, word)? {

meilisearch-core/src/store/synonyms.rs

@@ -1,4 +1,5 @@
 use heed::types::ByteSlice;
+use crate::database::MainT;
 use heed::Result as ZResult;
 use std::sync::Arc;
 
@@ -10,7 +11,7 @@ pub struct Synonyms {
 impl Synonyms {
     pub fn put_synonyms(
         self,
-        writer: &mut heed::RwTxn,
+        writer: &mut heed::RwTxn<MainT>,
         word: &[u8],
         synonyms: &fst::Set,
     ) -> ZResult<()> {
@@ -18,15 +19,19 @@ impl Synonyms {
         self.synonyms.put(writer, word, bytes)
     }
 
-    pub fn del_synonyms(self, writer: &mut heed::RwTxn, word: &[u8]) -> ZResult<bool> {
+    pub fn del_synonyms(self, writer: &mut heed::RwTxn<MainT>, word: &[u8]) -> ZResult<bool> {
         self.synonyms.delete(writer, word)
     }
 
-    pub fn synonyms(self, reader: &heed::RoTxn, word: &[u8]) -> ZResult<Option<fst::Set>> {
+    pub fn clear(self, writer: &mut heed::RwTxn<MainT>) -> ZResult<()> {
+        self.synonyms.clear(writer)
+    }
+
+    pub fn synonyms(self, reader: &heed::RoTxn<MainT>, word: &[u8]) -> ZResult<Option<fst::Set>> {
         match self.synonyms.get(reader, word)? {
             Some(bytes) => {
                 let len = bytes.len();
-                let bytes = Arc::from(bytes);
+                let bytes = Arc::new(bytes.to_owned());
                 let fst = fst::raw::Fst::from_shared_bytes(bytes, 0, len).unwrap();
                 Ok(Some(fst::Set::from(fst)))
             }

meilisearch-core/src/store/updates.rs

@@ -1,4 +1,5 @@
 use super::BEU64;
+use crate::database::UpdateT;
 use crate::update::Update;
 use heed::types::{OwnedType, SerdeJson};
 use heed::Result as ZResult;
@@ -10,7 +11,7 @@ pub struct Updates {
 
 impl Updates {
     // TODO do not trigger deserialize if possible
-    pub fn last_update_id(self, reader: &heed::RoTxn) -> ZResult<Option<(u64, Update)>> {
+    pub fn last_update(self, reader: &heed::RoTxn<UpdateT>) -> ZResult<Option<(u64, Update)>> {
         match self.updates.last(reader)? {
             Some((key, data)) => Ok(Some((key.get(), data))),
             None => Ok(None),
@@ -18,7 +19,7 @@ impl Updates {
     }
 
     // TODO do not trigger deserialize if possible
-    fn first_update_id(self, reader: &heed::RoTxn) -> ZResult<Option<(u64, Update)>> {
+    pub fn first_update(self, reader: &heed::RoTxn<UpdateT>) -> ZResult<Option<(u64, Update)>> {
         match self.updates.first(reader)? {
             Some((key, data)) => Ok(Some((key.get(), data))),
             None => Ok(None),
@@ -26,14 +27,14 @@ impl Updates {
     }
 
     // TODO do not trigger deserialize if possible
-    pub fn get(self, reader: &heed::RoTxn, update_id: u64) -> ZResult<Option<Update>> {
+    pub fn get(self, reader: &heed::RoTxn<UpdateT>, update_id: u64) -> ZResult<Option<Update>> {
         let update_id = BEU64::new(update_id);
         self.updates.get(reader, &update_id)
     }
 
     pub fn put_update(
         self,
-        writer: &mut heed::RwTxn,
+        writer: &mut heed::RwTxn<UpdateT>,
         update_id: u64,
         update: &Update,
     ) -> ZResult<()> {
@@ -42,8 +43,13 @@ impl Updates {
         self.updates.put(writer, &update_id, update)
     }
 
-    pub fn pop_front(self, writer: &mut heed::RwTxn) -> ZResult<Option<(u64, Update)>> {
-        match self.first_update_id(writer)? {
+    pub fn del_update(self, writer: &mut heed::RwTxn<UpdateT>, update_id: u64) -> ZResult<bool> {
+        let update_id = BEU64::new(update_id);
+        self.updates.delete(writer, &update_id)
+    }
+
+    pub fn pop_front(self, writer: &mut heed::RwTxn<UpdateT>) -> ZResult<Option<(u64, Update)>> {
+        match self.first_update(writer)? {
             Some((update_id, update)) => {
                 let key = BEU64::new(update_id);
                 self.updates.delete(writer, &key)?;
@@ -52,4 +58,8 @@ impl Updates {
             None => Ok(None),
         }
     }
+
+    pub fn clear(self, writer: &mut heed::RwTxn<UpdateT>) -> ZResult<()> {
+        self.updates.clear(writer)
+    }
 }

meilisearch-core/src/store/updates_results.rs

@@ -1,18 +1,18 @@
 use super::BEU64;
+use crate::database::UpdateT;
 use crate::update::ProcessedUpdateResult;
-use heed::types::{OwnedType, SerdeBincode};
+use heed::types::{OwnedType, SerdeJson};
 use heed::Result as ZResult;
 
 #[derive(Copy, Clone)]
 pub struct UpdatesResults {
-    pub(crate) updates_results:
-        heed::Database<OwnedType<BEU64>, SerdeBincode<ProcessedUpdateResult>>,
+    pub(crate) updates_results: heed::Database<OwnedType<BEU64>, SerdeJson<ProcessedUpdateResult>>,
 }
 
 impl UpdatesResults {
-    pub fn last_update_id(
+    pub fn last_update(
         self,
-        reader: &heed::RoTxn,
+        reader: &heed::RoTxn<UpdateT>,
     ) -> ZResult<Option<(u64, ProcessedUpdateResult)>> {
         match self.updates_results.last(reader)? {
             Some((key, data)) => Ok(Some((key.get(), data))),
@@ -22,7 +22,7 @@ impl UpdatesResults {
 
     pub fn put_update_result(
         self,
-        writer: &mut heed::RwTxn,
+        writer: &mut heed::RwTxn<UpdateT>,
         update_id: u64,
         update_result: &ProcessedUpdateResult,
     ) -> ZResult<()> {
@@ -32,10 +32,14 @@ impl UpdatesResults {
 
     pub fn update_result(
         self,
-        reader: &heed::RoTxn,
+        reader: &heed::RoTxn<UpdateT>,
         update_id: u64,
     ) -> ZResult<Option<ProcessedUpdateResult>> {
         let update_id = BEU64::new(update_id);
         self.updates_results.get(reader, &update_id)
     }
+
+    pub fn clear(self, writer: &mut heed::RwTxn<UpdateT>) -> ZResult<()> {
+        self.updates_results.clear(writer)
+    }
 }

meilisearch-core/src/update/clear_all.rs

@@ -1,8 +1,9 @@
+use crate::database::{MainT, UpdateT};
 use crate::update::{next_update_id, Update};
 use crate::{store, MResult, RankedMap};
 
 pub fn apply_clear_all(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<MainT>,
     main_store: store::Main,
     documents_fields_store: store::DocumentsFields,
     documents_fields_counts_store: store::DocumentsFieldsCounts,
@@ -21,12 +22,12 @@ pub fn apply_clear_all(
 }
 
 pub fn push_clear_all(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<UpdateT>,
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
 ) -> MResult<u64> {
     let last_update_id = next_update_id(writer, updates_store, updates_results_store)?;
-    let update = Update::ClearAll;
+    let update = Update::clear_all();
     updates_store.put_update(writer, last_update_id, &update)?;
 
     Ok(last_update_id)

meilisearch-core/src/update/customs_update.rs

@@ -1,9 +1,11 @@
-use crate::store;
-use crate::update::{next_update_id, Update};
 use heed::Result as ZResult;
 
+use crate::database::{MainT, UpdateT};
+use crate::store;
+use crate::update::{next_update_id, Update};
+
 pub fn apply_customs_update(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<MainT>,
     main_store: store::Main,
     customs: &[u8],
 ) -> ZResult<()> {
@@ -11,14 +13,14 @@ pub fn apply_customs_update(
 }
 
 pub fn push_customs_update(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<UpdateT>,
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
     customs: Vec<u8>,
 ) -> ZResult<u64> {
     let last_update_id = next_update_id(writer, updates_store, updates_results_store)?;
 
-    let update = Update::Customs(customs);
+    let update = Update::customs(customs);
     updates_store.put_update(writer, last_update_id, &update)?;
 
     Ok(last_update_id)

meilisearch-core/src/update/documents_addition.rs

@@ -2,10 +2,12 @@ use std::collections::HashMap;
 
 use fst::{set::OpBuilder, SetBuilder};
 use sdset::{duo::Union, SetOperation};
-use serde::Serialize;
+use serde::{Deserialize, Serialize};
 
+use crate::database::{MainT, UpdateT};
+use crate::database::{UpdateEvent, UpdateEventsEmitter};
 use crate::raw_indexer::RawIndexer;
-use crate::serde::{extract_document_id, serialize_value, Serializer};
+use crate::serde::{extract_document_id, serialize_value, Deserializer, Serializer};
 use crate::store;
 use crate::update::{apply_documents_deletion, next_update_id, Update};
 use crate::{Error, MResult, RankedMap};
@@ -13,21 +15,37 @@ use crate::{Error, MResult, RankedMap};
 pub struct DocumentsAddition<D> {
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
-    updates_notifier: crossbeam_channel::Sender<()>,
+    updates_notifier: UpdateEventsEmitter,
     documents: Vec<D>,
+    is_partial: bool,
 }
 
 impl<D> DocumentsAddition<D> {
     pub fn new(
         updates_store: store::Updates,
         updates_results_store: store::UpdatesResults,
-        updates_notifier: crossbeam_channel::Sender<()>,
+        updates_notifier: UpdateEventsEmitter,
     ) -> DocumentsAddition<D> {
         DocumentsAddition {
             updates_store,
             updates_results_store,
             updates_notifier,
             documents: Vec::new(),
+            is_partial: false,
+        }
+    }
+
+    pub fn new_partial(
+        updates_store: store::Updates,
+        updates_results_store: store::UpdatesResults,
+        updates_notifier: UpdateEventsEmitter,
+    ) -> DocumentsAddition<D> {
+        DocumentsAddition {
+            updates_store,
+            updates_results_store,
+            updates_notifier,
+            documents: Vec::new(),
+            is_partial: true,
         }
     }
 
@@ -35,16 +53,17 @@ impl<D> DocumentsAddition<D> {
         self.documents.push(document);
     }
 
-    pub fn finalize(self, writer: &mut heed::RwTxn) -> MResult<u64>
+    pub fn finalize(self, writer: &mut heed::RwTxn<UpdateT>) -> MResult<u64>
     where
         D: serde::Serialize,
     {
-        let _ = self.updates_notifier.send(());
+        let _ = self.updates_notifier.send(UpdateEvent::NewUpdate);
         let update_id = push_documents_addition(
             writer,
             self.updates_store,
             self.updates_results_store,
             self.documents,
+            self.is_partial,
         )?;
         Ok(update_id)
     }
@@ -57,10 +76,11 @@ impl<D> Extend<D> for DocumentsAddition<D> {
 }
 
 pub fn push_documents_addition<D: serde::Serialize>(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<UpdateT>,
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
     addition: Vec<D>,
+    is_partial: bool,
 ) -> MResult<u64> {
     let mut values = Vec::with_capacity(addition.len());
     for add in addition {
@@ -71,20 +91,25 @@ pub fn push_documents_addition<D: serde::Serialize>(
 
     let last_update_id = next_update_id(writer, updates_store, updates_results_store)?;
 
-    let update = Update::DocumentsAddition(values);
+    let update = if is_partial {
+        Update::documents_partial(values)
+    } else {
+        Update::documents_addition(values)
+    };
+
     updates_store.put_update(writer, last_update_id, &update)?;
 
     Ok(last_update_id)
 }
 
-pub fn apply_documents_addition(
-    writer: &mut heed::RwTxn,
+pub fn apply_documents_addition<'a, 'b>(
+    writer: &'a mut heed::RwTxn<'b, MainT>,
     main_store: store::Main,
     documents_fields_store: store::DocumentsFields,
     documents_fields_counts_store: store::DocumentsFieldsCounts,
     postings_lists_store: store::PostingsLists,
     docs_words_store: store::DocsWords,
-    addition: Vec<serde_json::Value>,
+    addition: Vec<HashMap<String, serde_json::Value>>,
 ) -> MResult<()> {
     let mut documents_additions = HashMap::new();
 
@@ -156,8 +181,104 @@ pub fn apply_documents_addition(
     )
 }
 
+pub fn apply_documents_partial_addition<'a, 'b>(
+    writer: &'a mut heed::RwTxn<'b, MainT>,
+    main_store: store::Main,
+    documents_fields_store: store::DocumentsFields,
+    documents_fields_counts_store: store::DocumentsFieldsCounts,
+    postings_lists_store: store::PostingsLists,
+    docs_words_store: store::DocsWords,
+    addition: Vec<HashMap<String, serde_json::Value>>,
+) -> MResult<()> {
+    let mut documents_additions = HashMap::new();
+
+    let schema = match main_store.schema(writer)? {
+        Some(schema) => schema,
+        None => return Err(Error::SchemaMissing),
+    };
+
+    let identifier = schema.identifier_name();
+
+    // 1. store documents ids for future deletion
+    for mut document in addition {
+        let document_id = match extract_document_id(identifier, &document)? {
+            Some(id) => id,
+            None => return Err(Error::MissingDocumentId),
+        };
+
+        let mut deserializer = Deserializer {
+            document_id,
+            reader: writer,
+            documents_fields: documents_fields_store,
+            schema: &schema,
+            attributes: None,
+        };
+
+        // retrieve the old document and
+        // update the new one with missing keys found in the old one
+        let result = Option::<HashMap<String, serde_json::Value>>::deserialize(&mut deserializer)?;
+        if let Some(old_document) = result {
+            for (key, value) in old_document {
+                document.entry(key).or_insert(value);
+            }
+        }
+
+        documents_additions.insert(document_id, document);
+    }
+
+    // 2. remove the documents posting lists
+    let number_of_inserted_documents = documents_additions.len();
+    let documents_ids = documents_additions.iter().map(|(id, _)| *id).collect();
+    apply_documents_deletion(
+        writer,
+        main_store,
+        documents_fields_store,
+        documents_fields_counts_store,
+        postings_lists_store,
+        docs_words_store,
+        documents_ids,
+    )?;
+
+    let mut ranked_map = match main_store.ranked_map(writer)? {
+        Some(ranked_map) => ranked_map,
+        None => RankedMap::default(),
+    };
+
+    let stop_words = match main_store.stop_words_fst(writer)? {
+        Some(stop_words) => stop_words,
+        None => fst::Set::default(),
+    };
+
+    // 3. index the documents fields in the stores
+    let mut indexer = RawIndexer::new(stop_words);
+
+    for (document_id, document) in documents_additions {
+        let serializer = Serializer {
+            txn: writer,
+            schema: &schema,
+            document_store: documents_fields_store,
+            document_fields_counts: documents_fields_counts_store,
+            indexer: &mut indexer,
+            ranked_map: &mut ranked_map,
+            document_id,
+        };
+
+        document.serialize(serializer)?;
+    }
+
+    write_documents_addition_index(
+        writer,
+        main_store,
+        postings_lists_store,
+        docs_words_store,
+        &ranked_map,
+        number_of_inserted_documents,
+        indexer,
+    )
+}
+
 pub fn reindex_all_documents(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<MainT>,
     main_store: store::Main,
     documents_fields_store: store::DocumentsFields,
     documents_fields_counts_store: store::DocumentsFieldsCounts,
@@ -234,7 +355,7 @@ pub fn reindex_all_documents(
 }
 
 pub fn write_documents_addition_index(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<MainT>,
     main_store: store::Main,
     postings_lists_store: store::PostingsLists,
     docs_words_store: store::DocsWords,

meilisearch-core/src/update/documents_deletion.rs

@@ -1,9 +1,11 @@
 use std::collections::{BTreeSet, HashMap, HashSet};
 
 use fst::{SetBuilder, Streamer};
-use meilidb_schema::Schema;
+use meilisearch_schema::Schema;
 use sdset::{duo::DifferenceByKey, SetBuf, SetOperation};
 
+use crate::database::{MainT, UpdateT};
+use crate::database::{UpdateEvent, UpdateEventsEmitter};
 use crate::serde::extract_document_id;
 use crate::store;
 use crate::update::{next_update_id, Update};
@@ -12,7 +14,7 @@ use crate::{DocumentId, Error, MResult, RankedMap};
 pub struct DocumentsDeletion {
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
-    updates_notifier: crossbeam_channel::Sender<()>,
+    updates_notifier: UpdateEventsEmitter,
     documents: Vec<DocumentId>,
 }
 
@@ -20,7 +22,7 @@ impl DocumentsDeletion {
     pub fn new(
         updates_store: store::Updates,
         updates_results_store: store::UpdatesResults,
-        updates_notifier: crossbeam_channel::Sender<()>,
+        updates_notifier: UpdateEventsEmitter,
     ) -> DocumentsDeletion {
         DocumentsDeletion {
             updates_store,
@@ -49,8 +51,8 @@ impl DocumentsDeletion {
         Ok(())
     }
 
-    pub fn finalize(self, writer: &mut heed::RwTxn) -> MResult<u64> {
-        let _ = self.updates_notifier.send(());
+    pub fn finalize(self, writer: &mut heed::RwTxn<UpdateT>) -> MResult<u64> {
+        let _ = self.updates_notifier.send(UpdateEvent::NewUpdate);
         let update_id = push_documents_deletion(
             writer,
             self.updates_store,
@@ -68,21 +70,21 @@ impl Extend<DocumentId> for DocumentsDeletion {
 }
 
 pub fn push_documents_deletion(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<UpdateT>,
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
     deletion: Vec<DocumentId>,
 ) -> MResult<u64> {
     let last_update_id = next_update_id(writer, updates_store, updates_results_store)?;
 
-    let update = Update::DocumentsDeletion(deletion);
+    let update = Update::documents_deletion(deletion);
     updates_store.put_update(writer, last_update_id, &update)?;
 
     Ok(last_update_id)
 }
 
 pub fn apply_documents_deletion(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<MainT>,
     main_store: store::Main,
     documents_fields_store: store::DocumentsFields,
     documents_fields_counts_store: store::DocumentsFieldsCounts,

meilisearch-core/src/update/mod.rs

@@ -10,7 +10,9 @@ mod synonyms_deletion;
 
 pub use self::clear_all::{apply_clear_all, push_clear_all};
 pub use self::customs_update::{apply_customs_update, push_customs_update};
-pub use self::documents_addition::{apply_documents_addition, DocumentsAddition};
+pub use self::documents_addition::{
+    apply_documents_addition, apply_documents_partial_addition, DocumentsAddition,
+};
 pub use self::documents_deletion::{apply_documents_deletion, DocumentsDeletion};
 pub use self::schema_update::{apply_schema_update, push_schema_update};
 pub use self::stop_words_addition::{apply_stop_words_addition, StopWordsAddition};
@@ -19,22 +21,103 @@ pub use self::synonyms_addition::{apply_synonyms_addition, SynonymsAddition};
 pub use self::synonyms_deletion::{apply_synonyms_deletion, SynonymsDeletion};
 
 use std::cmp;
-use std::collections::{BTreeMap, BTreeSet};
-use std::time::{Duration, Instant};
+use std::collections::{BTreeMap, BTreeSet, HashMap};
+use std::time::Instant;
 
+use chrono::{DateTime, Utc};
 use heed::Result as ZResult;
 use log::debug;
 use serde::{Deserialize, Serialize};
 
 use crate::{store, DocumentId, MResult};
-use meilidb_schema::Schema;
+use crate::database::{MainT, UpdateT};
+use meilisearch_schema::Schema;
 
 #[derive(Debug, Clone, Serialize, Deserialize)]
-pub enum Update {
+pub struct Update {
+    data: UpdateData,
+    enqueued_at: DateTime<Utc>,
+}
+
+impl Update {
+    fn clear_all() -> Update {
+        Update {
+            data: UpdateData::ClearAll,
+            enqueued_at: Utc::now(),
+        }
+    }
+
+    fn schema(data: Schema) -> Update {
+        Update {
+            data: UpdateData::Schema(data),
+            enqueued_at: Utc::now(),
+        }
+    }
+
+    fn customs(data: Vec<u8>) -> Update {
+        Update {
+            data: UpdateData::Customs(data),
+            enqueued_at: Utc::now(),
+        }
+    }
+
+    fn documents_addition(data: Vec<HashMap<String, serde_json::Value>>) -> Update {
+        Update {
+            data: UpdateData::DocumentsAddition(data),
+            enqueued_at: Utc::now(),
+        }
+    }
+
+    fn documents_partial(data: Vec<HashMap<String, serde_json::Value>>) -> Update {
+        Update {
+            data: UpdateData::DocumentsPartial(data),
+            enqueued_at: Utc::now(),
+        }
+    }
+
+    fn documents_deletion(data: Vec<DocumentId>) -> Update {
+        Update {
+            data: UpdateData::DocumentsDeletion(data),
+            enqueued_at: Utc::now(),
+        }
+    }
+
+    fn synonyms_addition(data: BTreeMap<String, Vec<String>>) -> Update {
+        Update {
+            data: UpdateData::SynonymsAddition(data),
+            enqueued_at: Utc::now(),
+        }
+    }
+
+    fn synonyms_deletion(data: BTreeMap<String, Option<Vec<String>>>) -> Update {
+        Update {
+            data: UpdateData::SynonymsDeletion(data),
+            enqueued_at: Utc::now(),
+        }
+    }
+
+    fn stop_words_addition(data: BTreeSet<String>) -> Update {
+        Update {
+            data: UpdateData::StopWordsAddition(data),
+            enqueued_at: Utc::now(),
+        }
+    }
+
+    fn stop_words_deletion(data: BTreeSet<String>) -> Update {
+        Update {
+            data: UpdateData::StopWordsDeletion(data),
+            enqueued_at: Utc::now(),
+        }
+    }
+}
+
+#[derive(Debug, Clone, Serialize, Deserialize)]
+pub enum UpdateData {
     ClearAll,
     Schema(Schema),
     Customs(Vec<u8>),
-    DocumentsAddition(Vec<serde_json::Value>),
+    DocumentsAddition(Vec<HashMap<String, serde_json::Value>>),
+    DocumentsPartial(Vec<HashMap<String, serde_json::Value>>),
     DocumentsDeletion(Vec<DocumentId>),
     SynonymsAddition(BTreeMap<String, Vec<String>>),
     SynonymsDeletion(BTreeMap<String, Option<Vec<String>>>),
@@ -42,30 +125,31 @@ pub enum Update {
     StopWordsDeletion(BTreeSet<String>),
 }
 
-impl Update {
+impl UpdateData {
     pub fn update_type(&self) -> UpdateType {
         match self {
-            Update::ClearAll => UpdateType::ClearAll,
-            Update::Schema(schema) => UpdateType::Schema {
-                schema: schema.clone(),
-            },
-            Update::Customs(_) => UpdateType::Customs,
-            Update::DocumentsAddition(addition) => UpdateType::DocumentsAddition {
+            UpdateData::ClearAll => UpdateType::ClearAll,
+            UpdateData::Schema(_) => UpdateType::Schema,
+            UpdateData::Customs(_) => UpdateType::Customs,
+            UpdateData::DocumentsAddition(addition) => UpdateType::DocumentsAddition {
                 number: addition.len(),
             },
-            Update::DocumentsDeletion(deletion) => UpdateType::DocumentsDeletion {
+            UpdateData::DocumentsPartial(addition) => UpdateType::DocumentsPartial {
+                number: addition.len(),
+            },
+            UpdateData::DocumentsDeletion(deletion) => UpdateType::DocumentsDeletion {
                 number: deletion.len(),
             },
-            Update::SynonymsAddition(addition) => UpdateType::SynonymsAddition {
+            UpdateData::SynonymsAddition(addition) => UpdateType::SynonymsAddition {
                 number: addition.len(),
             },
-            Update::SynonymsDeletion(deletion) => UpdateType::SynonymsDeletion {
+            UpdateData::SynonymsDeletion(deletion) => UpdateType::SynonymsDeletion {
                 number: deletion.len(),
             },
-            Update::StopWordsAddition(addition) => UpdateType::StopWordsAddition {
+            UpdateData::StopWordsAddition(addition) => UpdateType::StopWordsAddition {
                 number: addition.len(),
             },
-            Update::StopWordsDeletion(deletion) => UpdateType::StopWordsDeletion {
+            UpdateData::StopWordsDeletion(deletion) => UpdateType::StopWordsDeletion {
                 number: deletion.len(),
             },
         }
@@ -73,11 +157,13 @@ impl Update {
 }
 
 #[derive(Debug, Clone, Serialize, Deserialize)]
+#[serde(tag = "name")]
 pub enum UpdateType {
     ClearAll,
-    Schema { schema: Schema },
+    Schema,
     Customs,
     DocumentsAddition { number: usize },
+    DocumentsPartial { number: usize },
     DocumentsDeletion { number: usize },
     SynonymsAddition { number: usize },
     SynonymsDeletion { number: usize },
@@ -85,83 +171,88 @@ pub enum UpdateType {
     StopWordsDeletion { number: usize },
 }
 
-#[derive(Debug, Clone, Serialize, Deserialize)]
-pub struct DetailedDuration {
-    pub main: Duration,
-}
-
 #[derive(Debug, Clone, Serialize, Deserialize)]
 pub struct ProcessedUpdateResult {
     pub update_id: u64,
+    #[serde(rename = "type")]
     pub update_type: UpdateType,
-    pub result: Result<(), String>,
-    pub detailed_duration: DetailedDuration,
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub error: Option<String>,
+    pub duration: f64, // in seconds
+    pub enqueued_at: DateTime<Utc>,
+    pub processed_at: DateTime<Utc>,
 }
 
 #[derive(Debug, Clone, Serialize, Deserialize)]
 pub struct EnqueuedUpdateResult {
     pub update_id: u64,
     pub update_type: UpdateType,
+    pub enqueued_at: DateTime<Utc>,
 }
 
 #[derive(Debug, Clone, Serialize, Deserialize)]
+#[serde(rename_all = "camelCase", tag = "status")]
 pub enum UpdateStatus {
-    Enqueued(EnqueuedUpdateResult),
-    Processed(ProcessedUpdateResult),
-    Unknown,
+    Enqueued {
+        #[serde(flatten)]
+        content: EnqueuedUpdateResult,
+    },
+    Processed {
+        #[serde(flatten)]
+        content: ProcessedUpdateResult,
+    },
 }
 
 pub fn update_status(
-    reader: &heed::RoTxn,
+    update_reader: &heed::RoTxn<UpdateT>,
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
     update_id: u64,
-) -> MResult<UpdateStatus> {
-    match updates_results_store.update_result(reader, update_id)? {
-        Some(result) => Ok(UpdateStatus::Processed(result)),
-        None => {
-            if let Some(update) = updates_store.get(reader, update_id)? {
-                Ok(UpdateStatus::Enqueued(EnqueuedUpdateResult {
+) -> MResult<Option<UpdateStatus>> {
+    match updates_results_store.update_result(update_reader, update_id)? {
+        Some(result) => Ok(Some(UpdateStatus::Processed { content: result })),
+        None => match updates_store.get(update_reader, update_id)? {
+            Some(update) => Ok(Some(UpdateStatus::Enqueued {
+                content: EnqueuedUpdateResult {
                     update_id,
-                    update_type: update.update_type(),
-                }))
-            } else {
-                Ok(UpdateStatus::Unknown)
-            }
-        }
+                    update_type: update.data.update_type(),
+                    enqueued_at: update.enqueued_at,
+                },
+            })),
+            None => Ok(None),
+        },
     }
 }
 
 pub fn next_update_id(
-    writer: &mut heed::RwTxn,
+    update_writer: &mut heed::RwTxn<UpdateT>,
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
 ) -> ZResult<u64> {
-    let last_update_id = updates_store.last_update_id(writer)?;
-    let last_update_id = last_update_id.map(|(n, _)| n);
+    let last_update = updates_store.last_update(update_writer)?;
+    let last_update = last_update.map(|(n, _)| n);
 
-    let last_update_results_id = updates_results_store.last_update_id(writer)?;
+    let last_update_results_id = updates_results_store.last_update(update_writer)?;
     let last_update_results_id = last_update_results_id.map(|(n, _)| n);
 
-    let max_update_id = cmp::max(last_update_id, last_update_results_id);
+    let max_update_id = cmp::max(last_update, last_update_results_id);
     let new_update_id = max_update_id.map_or(0, |n| n + 1);
 
     Ok(new_update_id)
 }
 
-pub fn update_task(
-    writer: &mut heed::RwTxn,
-    index: store::Index,
-) -> MResult<Option<ProcessedUpdateResult>> {
-    let (update_id, update) = match index.updates.pop_front(writer)? {
-        Some(value) => value,
-        None => return Ok(None),
-    };
-
+pub fn update_task<'a, 'b>(
+    writer: &'a mut heed::RwTxn<'b, MainT>,
+    index: &store::Index,
+    update_id: u64,
+    update: Update,
+) -> MResult<ProcessedUpdateResult> {
     debug!("Processing update number {}", update_id);
 
-    let (update_type, result, duration) = match update {
-        Update::ClearAll => {
+    let Update { enqueued_at, data } = update;
+
+    let (update_type, result, duration) = match data {
+        UpdateData::ClearAll => {
             let start = Instant::now();
 
             let update_type = UpdateType::ClearAll;
@@ -176,12 +267,10 @@ pub fn update_task(
 
             (update_type, result, start.elapsed())
         }
-        Update::Schema(schema) => {
+        UpdateData::Schema(schema) => {
             let start = Instant::now();
 
-            let update_type = UpdateType::Schema {
-                schema: schema.clone(),
-            };
+            let update_type = UpdateType::Schema;
             let result = apply_schema_update(
                 writer,
                 &schema,
@@ -194,7 +283,7 @@ pub fn update_task(
 
             (update_type, result, start.elapsed())
         }
-        Update::Customs(customs) => {
+        UpdateData::Customs(customs) => {
             let start = Instant::now();
 
             let update_type = UpdateType::Customs;
@@ -202,7 +291,7 @@ pub fn update_task(
 
             (update_type, result, start.elapsed())
         }
-        Update::DocumentsAddition(documents) => {
+        UpdateData::DocumentsAddition(documents) => {
             let start = Instant::now();
 
             let update_type = UpdateType::DocumentsAddition {
@@ -221,7 +310,26 @@ pub fn update_task(
 
             (update_type, result, start.elapsed())
         }
-        Update::DocumentsDeletion(documents) => {
+        UpdateData::DocumentsPartial(documents) => {
+            let start = Instant::now();
+
+            let update_type = UpdateType::DocumentsPartial {
+                number: documents.len(),
+            };
+
+            let result = apply_documents_partial_addition(
+                writer,
+                index.main,
+                index.documents_fields,
+                index.documents_fields_counts,
+                index.postings_lists,
+                index.docs_words,
+                documents,
+            );
+
+            (update_type, result, start.elapsed())
+        }
+        UpdateData::DocumentsDeletion(documents) => {
             let start = Instant::now();
 
             let update_type = UpdateType::DocumentsDeletion {
@@ -240,7 +348,7 @@ pub fn update_task(
 
             (update_type, result, start.elapsed())
         }
-        Update::SynonymsAddition(synonyms) => {
+        UpdateData::SynonymsAddition(synonyms) => {
             let start = Instant::now();
 
             let update_type = UpdateType::SynonymsAddition {
@@ -251,7 +359,7 @@ pub fn update_task(
 
             (update_type, result, start.elapsed())
         }
-        Update::SynonymsDeletion(synonyms) => {
+        UpdateData::SynonymsDeletion(synonyms) => {
             let start = Instant::now();
 
             let update_type = UpdateType::SynonymsDeletion {
@@ -262,7 +370,7 @@ pub fn update_task(
 
             (update_type, result, start.elapsed())
         }
-        Update::StopWordsAddition(stop_words) => {
+        UpdateData::StopWordsAddition(stop_words) => {
             let start = Instant::now();
 
             let update_type = UpdateType::StopWordsAddition {
@@ -274,7 +382,7 @@ pub fn update_task(
 
             (update_type, result, start.elapsed())
         }
-        Update::StopWordsDeletion(stop_words) => {
+        UpdateData::StopWordsDeletion(stop_words) => {
             let start = Instant::now();
 
             let update_type = UpdateType::StopWordsDeletion {
@@ -300,17 +408,14 @@ pub fn update_task(
         update_id, update_type, result
     );
 
-    let detailed_duration = DetailedDuration { main: duration };
     let status = ProcessedUpdateResult {
         update_id,
         update_type,
-        result: result.map_err(|e| e.to_string()),
-        detailed_duration,
+        error: result.map_err(|e| e.to_string()).err(),
+        duration: duration.as_secs_f64(),
+        enqueued_at,
+        processed_at: Utc::now(),
     };
 
-    index
-        .updates_results
-        .put_update_result(writer, update_id, &status)?;
-
-    Ok(Some(status))
+    Ok(status)
 }

meilisearch-core/src/update/schema_update.rs

@@ -1,11 +1,12 @@
-use meilidb_schema::{Diff, Schema};
+use meilisearch_schema::{Diff, Schema};
 
+use crate::database::{MainT, UpdateT};
 use crate::update::documents_addition::reindex_all_documents;
 use crate::update::{next_update_id, Update};
 use crate::{error::UnsupportedOperation, store, MResult};
 
 pub fn apply_schema_update(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<MainT>,
     new_schema: &Schema,
     main_store: store::Main,
     documents_fields_store: store::DocumentsFields,
@@ -14,14 +15,14 @@ pub fn apply_schema_update(
     docs_words_store: store::DocsWords,
 ) -> MResult<()> {
     use UnsupportedOperation::{
-        CannotIntroduceNewSchemaAttribute, CannotRemoveSchemaAttribute,
+        CanOnlyIntroduceNewSchemaAttributesAtEnd, CannotRemoveSchemaAttribute,
         CannotReorderSchemaAttribute, CannotUpdateSchemaIdentifier,
     };
 
     let mut need_full_reindexing = false;
 
     if let Some(old_schema) = main_store.schema(writer)? {
-        for diff in meilidb_schema::diff(&old_schema, new_schema) {
+        for diff in meilisearch_schema::diff(&old_schema, new_schema) {
             match diff {
                 Diff::IdentChange { .. } => return Err(CannotUpdateSchemaIdentifier.into()),
                 Diff::AttrMove { .. } => return Err(CannotReorderSchemaAttribute.into()),
@@ -33,7 +34,12 @@ pub fn apply_schema_update(
                         need_full_reindexing = true;
                     }
                 }
-                Diff::NewAttr { .. } => return Err(CannotIntroduceNewSchemaAttribute.into()),
+                Diff::NewAttr { pos, .. } => {
+                    // new attribute not at the end of the schema
+                    if pos < old_schema.number_of_attributes() {
+                        return Err(CanOnlyIntroduceNewSchemaAttributesAtEnd.into());
+                    }
+                }
                 Diff::RemovedAttr { .. } => return Err(CannotRemoveSchemaAttribute.into()),
             }
         }
@@ -56,14 +62,14 @@ pub fn apply_schema_update(
 }
 
 pub fn push_schema_update(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<UpdateT>,
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
     schema: Schema,
 ) -> MResult<u64> {
     let last_update_id = next_update_id(writer, updates_store, updates_results_store)?;
 
-    let update = Update::Schema(schema);
+    let update = Update::schema(schema);
     updates_store.put_update(writer, last_update_id, &update)?;
 
     Ok(last_update_id)

meilisearch-core/src/update/stop_words_addition.rs

@@ -2,14 +2,16 @@ use std::collections::BTreeSet;
 
 use fst::{set::OpBuilder, SetBuilder};
 
+use crate::database::{MainT, UpdateT};
 use crate::automaton::normalize_str;
+use crate::database::{UpdateEvent, UpdateEventsEmitter};
 use crate::update::{next_update_id, Update};
 use crate::{store, MResult};
 
 pub struct StopWordsAddition {
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
-    updates_notifier: crossbeam_channel::Sender<()>,
+    updates_notifier: UpdateEventsEmitter,
     stop_words: BTreeSet<String>,
 }
 
@@ -17,7 +19,7 @@ impl StopWordsAddition {
     pub fn new(
         updates_store: store::Updates,
         updates_results_store: store::UpdatesResults,
-        updates_notifier: crossbeam_channel::Sender<()>,
+        updates_notifier: UpdateEventsEmitter,
     ) -> StopWordsAddition {
         StopWordsAddition {
             updates_store,
@@ -32,8 +34,8 @@ impl StopWordsAddition {
         self.stop_words.insert(stop_word);
     }
 
-    pub fn finalize(self, writer: &mut heed::RwTxn) -> MResult<u64> {
-        let _ = self.updates_notifier.send(());
+    pub fn finalize(self, writer: &mut heed::RwTxn<UpdateT>) -> MResult<u64> {
+        let _ = self.updates_notifier.send(UpdateEvent::NewUpdate);
         let update_id = push_stop_words_addition(
             writer,
             self.updates_store,
@@ -45,21 +47,21 @@ impl StopWordsAddition {
 }
 
 pub fn push_stop_words_addition(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<UpdateT>,
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
     addition: BTreeSet<String>,
 ) -> MResult<u64> {
     let last_update_id = next_update_id(writer, updates_store, updates_results_store)?;
 
-    let update = Update::StopWordsAddition(addition);
+    let update = Update::stop_words_addition(addition);
     updates_store.put_update(writer, last_update_id, &update)?;
 
     Ok(last_update_id)
 }
 
 pub fn apply_stop_words_addition(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<MainT>,
     main_store: store::Main,
     postings_lists_store: store::PostingsLists,
     addition: BTreeSet<String>,

meilisearch-core/src/update/stop_words_deletion.rs

@@ -2,7 +2,9 @@ use std::collections::BTreeSet;
 
 use fst::{set::OpBuilder, SetBuilder};
 
+use crate::database::{MainT, UpdateT};
 use crate::automaton::normalize_str;
+use crate::database::{UpdateEvent, UpdateEventsEmitter};
 use crate::update::documents_addition::reindex_all_documents;
 use crate::update::{next_update_id, Update};
 use crate::{store, MResult};
@@ -10,7 +12,7 @@ use crate::{store, MResult};
 pub struct StopWordsDeletion {
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
-    updates_notifier: crossbeam_channel::Sender<()>,
+    updates_notifier: UpdateEventsEmitter,
     stop_words: BTreeSet<String>,
 }
 
@@ -18,7 +20,7 @@ impl StopWordsDeletion {
     pub fn new(
         updates_store: store::Updates,
         updates_results_store: store::UpdatesResults,
-        updates_notifier: crossbeam_channel::Sender<()>,
+        updates_notifier: UpdateEventsEmitter,
     ) -> StopWordsDeletion {
         StopWordsDeletion {
             updates_store,
@@ -33,8 +35,8 @@ impl StopWordsDeletion {
         self.stop_words.insert(stop_word);
     }
 
-    pub fn finalize(self, writer: &mut heed::RwTxn) -> MResult<u64> {
-        let _ = self.updates_notifier.send(());
+    pub fn finalize(self, writer: &mut heed::RwTxn<UpdateT>) -> MResult<u64> {
+        let _ = self.updates_notifier.send(UpdateEvent::NewUpdate);
         let update_id = push_stop_words_deletion(
             writer,
             self.updates_store,
@@ -46,21 +48,21 @@ impl StopWordsDeletion {
 }
 
 pub fn push_stop_words_deletion(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<UpdateT>,
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
     deletion: BTreeSet<String>,
 ) -> MResult<u64> {
     let last_update_id = next_update_id(writer, updates_store, updates_results_store)?;
 
-    let update = Update::StopWordsDeletion(deletion);
+    let update = Update::stop_words_deletion(deletion);
     updates_store.put_update(writer, last_update_id, &update)?;
 
     Ok(last_update_id)
 }
 
 pub fn apply_stop_words_deletion(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<MainT>,
     main_store: store::Main,
     documents_fields_store: store::DocumentsFields,
     documents_fields_counts_store: store::DocumentsFieldsCounts,

meilisearch-core/src/update/synonyms_addition.rs

@@ -3,14 +3,16 @@ use std::collections::BTreeMap;
 use fst::{set::OpBuilder, SetBuilder};
 use sdset::SetBuf;
 
+use crate::database::{MainT, UpdateT};
 use crate::automaton::normalize_str;
+use crate::database::{UpdateEvent, UpdateEventsEmitter};
 use crate::update::{next_update_id, Update};
 use crate::{store, MResult};
 
 pub struct SynonymsAddition {
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
-    updates_notifier: crossbeam_channel::Sender<()>,
+    updates_notifier: UpdateEventsEmitter,
     synonyms: BTreeMap<String, Vec<String>>,
 }
 
@@ -18,7 +20,7 @@ impl SynonymsAddition {
     pub fn new(
         updates_store: store::Updates,
         updates_results_store: store::UpdatesResults,
-        updates_notifier: crossbeam_channel::Sender<()>,
+        updates_notifier: UpdateEventsEmitter,
     ) -> SynonymsAddition {
         SynonymsAddition {
             updates_store,
@@ -42,8 +44,8 @@ impl SynonymsAddition {
             .extend(alternatives);
     }
 
-    pub fn finalize(self, writer: &mut heed::RwTxn) -> MResult<u64> {
-        let _ = self.updates_notifier.send(());
+    pub fn finalize(self, writer: &mut heed::RwTxn<UpdateT>) -> MResult<u64> {
+        let _ = self.updates_notifier.send(UpdateEvent::NewUpdate);
         let update_id = push_synonyms_addition(
             writer,
             self.updates_store,
@@ -55,21 +57,21 @@ impl SynonymsAddition {
 }
 
 pub fn push_synonyms_addition(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<UpdateT>,
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
     addition: BTreeMap<String, Vec<String>>,
 ) -> MResult<u64> {
     let last_update_id = next_update_id(writer, updates_store, updates_results_store)?;
 
-    let update = Update::SynonymsAddition(addition);
+    let update = Update::synonyms_addition(addition);
     updates_store.put_update(writer, last_update_id, &update)?;
 
     Ok(last_update_id)
 }
 
 pub fn apply_synonyms_addition(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<MainT>,
     main_store: store::Main,
     synonyms_store: store::Synonyms,
     addition: BTreeMap<String, Vec<String>>,

meilisearch-core/src/update/synonyms_deletion.rs

@@ -4,14 +4,16 @@ use std::iter::FromIterator;
 use fst::{set::OpBuilder, SetBuilder};
 use sdset::SetBuf;
 
+use crate::database::{MainT, UpdateT};
 use crate::automaton::normalize_str;
+use crate::database::{UpdateEvent, UpdateEventsEmitter};
 use crate::update::{next_update_id, Update};
 use crate::{store, MResult};
 
 pub struct SynonymsDeletion {
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
-    updates_notifier: crossbeam_channel::Sender<()>,
+    updates_notifier: UpdateEventsEmitter,
     synonyms: BTreeMap<String, Option<Vec<String>>>,
 }
 
@@ -19,7 +21,7 @@ impl SynonymsDeletion {
     pub fn new(
         updates_store: store::Updates,
         updates_results_store: store::UpdatesResults,
-        updates_notifier: crossbeam_channel::Sender<()>,
+        updates_notifier: UpdateEventsEmitter,
     ) -> SynonymsDeletion {
         SynonymsDeletion {
             updates_store,
@@ -49,8 +51,8 @@ impl SynonymsDeletion {
         }
     }
 
-    pub fn finalize(self, writer: &mut heed::RwTxn) -> MResult<u64> {
-        let _ = self.updates_notifier.send(());
+    pub fn finalize(self, writer: &mut heed::RwTxn<UpdateT>) -> MResult<u64> {
+        let _ = self.updates_notifier.send(UpdateEvent::NewUpdate);
         let update_id = push_synonyms_deletion(
             writer,
             self.updates_store,
@@ -62,21 +64,21 @@ impl SynonymsDeletion {
 }
 
 pub fn push_synonyms_deletion(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<UpdateT>,
     updates_store: store::Updates,
     updates_results_store: store::UpdatesResults,
     deletion: BTreeMap<String, Option<Vec<String>>>,
 ) -> MResult<u64> {
     let last_update_id = next_update_id(writer, updates_store, updates_results_store)?;
 
-    let update = Update::SynonymsDeletion(deletion);
+    let update = Update::synonyms_deletion(deletion);
     updates_store.put_update(writer, last_update_id, &update)?;
 
     Ok(last_update_id)
 }
 
 pub fn apply_synonyms_deletion(
-    writer: &mut heed::RwTxn,
+    writer: &mut heed::RwTxn<MainT>,
     main_store: store::Main,
     synonyms_store: store::Synonyms,
     deletion: BTreeMap<String, Option<Vec<String>>>,

meilisearch-http/Cargo.toml

@@ -0,0 +1,64 @@
+[package]
+name = "meilisearch-http"
+version = "0.8.1"
+authors = [
+    "Quentin de Quelen <quentin@dequelen.me>",
+    "Clément Renault <clement@meilisearch.com>",
+]
+edition = "2018"
+
+[[bin]]
+name = "meilisearch"
+path = "src/main.rs"
+
+[dependencies]
+bincode = "1.2.0"
+chrono = { version = "0.4.9", features = ["serde"] }
+crossbeam-channel = "0.4.0"
+env_logger = "0.7.1"
+heed = "0.6.0"
+http = "0.1.19"
+indexmap = { version = "1.3.0", features = ["serde-1"] }
+isahc = "0.7.6"
+log = "0.4.8"
+main_error = "0.1.0"
+meilisearch-core = { path = "../meilisearch-core", version = "0.8.1" }
+meilisearch-schema = { path = "../meilisearch-schema", version = "0.8.1" }
+pretty-bytes = "0.2.2"
+rand = "0.7.2"
+rayon = "1.2.0"
+serde = { version = "1.0.101", features = ["derive"] }
+serde_json = { version = "1.0.41", features = ["preserve_order"] }
+serde_qs = "0.5.1"
+siphasher = "0.3.1"
+structopt = "0.3.3"
+sysinfo = "0.9.5"
+walkdir = "2.2.9"
+whoami = "0.6"
+
+[dependencies.async-compression]
+default-features = false
+features = ["stream", "gzip", "zlib", "brotli", "zstd"]
+version = "=0.1.0-alpha.7"
+
+[dependencies.tide]
+git = "https://github.com/rustasync/tide"
+rev = "e77709370bb24cf776fe6da902467c35131535b1"
+
+[dependencies.tide-log]
+git = "https://github.com/rustasync/tide"
+rev = "e77709370bb24cf776fe6da902467c35131535b1"
+
+[dependencies.tide-slog]
+git = "https://github.com/rustasync/tide"
+rev = "e77709370bb24cf776fe6da902467c35131535b1"
+
+[dependencies.tide-compression]
+git = "https://github.com/rustasync/tide"
+rev = "e77709370bb24cf776fe6da902467c35131535b1"
+
+[build-dependencies]
+vergen = "3.0.4"
+
+[target.'cfg(unix)'.dependencies]
+jemallocator = "0.3.2"

meilisearch-http/build.rs

@@ -0,0 +1,10 @@
+use vergen::{generate_cargo_keys, ConstantsFlags};
+
+fn main() {
+    // Setup the flags, toggling off the 'SEMVER_FROM_CARGO_PKG' flag
+    let mut flags = ConstantsFlags::all();
+    flags.toggle(ConstantsFlags::SEMVER_FROM_CARGO_PKG);
+
+    // Generate the 'cargo:' key output
+    generate_cargo_keys(ConstantsFlags::all()).expect("Unable to generate the cargo keys!");
+}

meilisearch-http/src/analytics.rs

@@ -0,0 +1,73 @@
+use std::hash::{Hash, Hasher};
+use std::thread;
+use std::time::{Duration, SystemTime, UNIX_EPOCH};
+
+use log::error;
+use serde::Serialize;
+use serde_qs as qs;
+use siphasher::sip::SipHasher;
+
+const AMPLITUDE_API_KEY: &str = "f7fba398780e06d8fe6666a9be7e3d47";
+
+#[derive(Debug, Serialize)]
+struct Event<'a> {
+    user_id: &'a str,
+    event_type: &'a str,
+    device_id: &'a str,
+    time: u64,
+}
+
+#[derive(Debug, Serialize)]
+struct AmplitudeRequest<'a> {
+    api_key: &'a str,
+    event: &'a str,
+}
+
+pub fn analytics_sender() {
+    let username = whoami::username();
+    let hostname = whoami::hostname();
+    let platform = whoami::platform();
+
+    let uid = username + &hostname + &platform.to_string();
+
+    let mut hasher = SipHasher::new();
+    uid.hash(&mut hasher);
+    let hash = hasher.finish();
+
+    let uid = format!("{:X}", hash);
+    let platform = platform.to_string();
+
+    loop {
+        let n = SystemTime::now().duration_since(UNIX_EPOCH).unwrap();
+        let user_id = &uid;
+        let device_id = &platform;
+        let time = n.as_secs();
+        let event_type = "runtime_tick";
+
+        let event = Event {
+            user_id,
+            event_type,
+            device_id,
+            time,
+        };
+        let event = serde_json::to_string(&event).unwrap();
+
+        let request = AmplitudeRequest {
+            api_key: AMPLITUDE_API_KEY,
+            event: &event,
+        };
+
+        let body = qs::to_string(&request).unwrap();
+        match isahc::post("https://api.amplitude.com/httpapi", body) {
+            Ok(response) => {
+                if !response.status().is_success() {
+                    let body = response.into_body().text().unwrap();
+                    error!("Unsuccessful call to Amplitude: {}", body);
+                }
+            }
+            Err(e) => error!("Error while sending a request to Amplitude: {}", e),
+        }
+
+        thread::sleep(Duration::from_secs(86_400)) // one day
+    }
+}

meilisearch-http/src/data.rs

@@ -0,0 +1,129 @@
+use std::collections::HashMap;
+use std::ops::Deref;
+use std::sync::Arc;
+
+use chrono::{DateTime, Utc};
+use heed::types::{SerdeBincode, Str};
+use log::error;
+use meilisearch_core::{Database, MainT, UpdateT, Error as MError, MResult};
+use sysinfo::Pid;
+
+use crate::option::Opt;
+use crate::routes::index::index_update_callback;
+
+const LAST_UPDATE_KEY: &str = "last-update";
+
+type SerdeDatetime = SerdeBincode<DateTime<Utc>>;
+
+#[derive(Clone)]
+pub struct Data {
+    inner: Arc<DataInner>,
+}
+
+impl Deref for Data {
+    type Target = DataInner;
+
+    fn deref(&self) -> &Self::Target {
+        &self.inner
+    }
+}
+
+#[derive(Clone)]
+pub struct DataInner {
+    pub db: Arc<Database>,
+    pub db_path: String,
+    pub api_key: Option<String>,
+    pub server_pid: Pid,
+}
+
+impl DataInner {
+    pub fn is_indexing(&self, reader: &heed::RoTxn<UpdateT>, index: &str) -> MResult<Option<bool>> {
+        match self.db.open_index(&index) {
+            Some(index) => index.current_update_id(&reader).map(|u| Some(u.is_some())),
+            None => Ok(None),
+        }
+    }
+
+    pub fn last_update(&self, reader: &heed::RoTxn<MainT>) -> MResult<Option<DateTime<Utc>>> {
+        match self
+            .db
+            .common_store()
+            .get::<_, Str, SerdeDatetime>(reader, LAST_UPDATE_KEY)?
+        {
+            Some(datetime) => Ok(Some(datetime)),
+            None => Ok(None),
+        }
+    }
+
+    pub fn set_last_update(&self, writer: &mut heed::RwTxn<MainT>) -> MResult<()> {
+        self.db
+            .common_store()
+            .put::<_, Str, SerdeDatetime>(writer, LAST_UPDATE_KEY, &Utc::now())
+            .map_err(Into::into)
+    }
+
+    pub fn compute_stats(&self, writer: &mut heed::RwTxn<MainT>, index_uid: &str) -> MResult<()> {
+        let index = match self.db.open_index(&index_uid) {
+            Some(index) => index,
+            None => {
+                error!("Impossible to retrieve index {}", index_uid);
+                return Ok(());
+            }
+        };
+
+        let schema = match index.main.schema(&writer)? {
+            Some(schema) => schema,
+            None => return Ok(()),
+        };
+
+        let all_documents_fields = index
+            .documents_fields_counts
+            .all_documents_fields_counts(&writer)?;
+
+        // count fields frequencies
+        let mut fields_frequency = HashMap::<_, usize>::new();
+        for result in all_documents_fields {
+            let (_, attr, _) = result?;
+            *fields_frequency.entry(attr).or_default() += 1;
+        }
+
+        // convert attributes to their names
+        let frequency: HashMap<_, _> = fields_frequency
+            .into_iter()
+            .map(|(a, c)| (schema.attribute_name(a).to_owned(), c))
+            .collect();
+
+        index
+            .main
+            .put_fields_frequency(writer, &frequency)
+            .map_err(MError::Zlmdb)
+    }
+}
+
+impl Data {
+    pub fn new(opt: Opt) -> Data {
+        let db_path = opt.db_path.clone();
+        let api_key = opt.api_key.clone();
+        let server_pid = sysinfo::get_current_pid().unwrap();
+
+        let db = Arc::new(Database::open_or_create(opt.db_path.clone()).unwrap());
+
+        let inner_data = DataInner {
+            db: db.clone(),
+            db_path,
+            api_key,
+            server_pid,
+        };
+
+        let data = Data {
+            inner: Arc::new(inner_data),
+        };
+
+        let callback_context = data.clone();
+        db.set_update_callback(Box::new(move |index_uid, status| {
+            index_update_callback(&index_uid, &callback_context, status);
+        }));
+
+        data
+    }
+}

meilisearch-http/src/error.rs

@@ -0,0 +1,126 @@
+use std::fmt::Display;
+
+use http::status::StatusCode;
+use log::{error, warn};
+use serde::{Deserialize, Serialize};
+use tide::response::IntoResponse;
+use tide::Response;
+
+pub type SResult<T> = Result<T, ResponseError>;
+
+pub enum ResponseError {
+    Internal(String),
+    BadRequest(String),
+    InvalidToken(String),
+    NotFound(String),
+    IndexNotFound(String),
+    DocumentNotFound(String),
+    MissingHeader(String),
+    BadParameter(String, String),
+    OpenIndex(String),
+    CreateIndex(String),
+    Maintenance,
+}
+
+impl ResponseError {
+    pub fn internal(message: impl Display) -> ResponseError {
+        ResponseError::Internal(message.to_string())
+    }
+
+    pub fn bad_request(message: impl Display) -> ResponseError {
+        ResponseError::BadRequest(message.to_string())
+    }
+
+    pub fn invalid_token(message: impl Display) -> ResponseError {
+        ResponseError::InvalidToken(message.to_string())
+    }
+
+    pub fn not_found(message: impl Display) -> ResponseError {
+        ResponseError::NotFound(message.to_string())
+    }
+
+    pub fn index_not_found(message: impl Display) -> ResponseError {
+        ResponseError::IndexNotFound(message.to_string())
+    }
+
+    pub fn document_not_found(message: impl Display) -> ResponseError {
+        ResponseError::DocumentNotFound(message.to_string())
+    }
+
+    pub fn missing_header(message: impl Display) -> ResponseError {
+        ResponseError::MissingHeader(message.to_string())
+    }
+
+    pub fn bad_parameter(name: impl Display, message: impl Display) -> ResponseError {
+        ResponseError::BadParameter(name.to_string(), message.to_string())
+    }
+
+    pub fn open_index(message: impl Display) -> ResponseError {
+        ResponseError::OpenIndex(message.to_string())
+    }
+
+    pub fn create_index(message: impl Display) -> ResponseError {
+        ResponseError::CreateIndex(message.to_string())
+    }
+}
+
+impl IntoResponse for ResponseError {
+    fn into_response(self) -> Response {
+        match self {
+            ResponseError::Internal(err) => {
+                error!("internal server error: {}", err);
+                error(
+                    String::from("Internal server error"),
+                    StatusCode::INTERNAL_SERVER_ERROR,
+                )
+            }
+            ResponseError::BadRequest(err) => {
+                warn!("bad request: {}", err);
+                error(err, StatusCode::BAD_REQUEST)
+            }
+            ResponseError::InvalidToken(err) => {
+                error(format!("Invalid Token: {}", err), StatusCode::FORBIDDEN)
+            }
+            ResponseError::NotFound(err) => error(err, StatusCode::NOT_FOUND),
+            ResponseError::IndexNotFound(index) => {
+                error(format!("Index {} not found", index), StatusCode::NOT_FOUND)
+            }
+            ResponseError::DocumentNotFound(id) => error(
+                format!("Document with id {} not found", id),
+                StatusCode::NOT_FOUND,
+            ),
+            ResponseError::MissingHeader(header) => error(
+                format!("Header {} is missing", header),
+                StatusCode::UNAUTHORIZED,
+            ),
+            ResponseError::BadParameter(param, e) => error(
+                format!("Url parameter {} error: {}", param, e),
+                StatusCode::BAD_REQUEST,
+            ),
+            ResponseError::CreateIndex(err) => error(
+                format!("Impossible to create index; {}", err),
+                StatusCode::BAD_REQUEST,
+            ),
+            ResponseError::OpenIndex(err) => error(
+                format!("Impossible to open index; {}", err),
+                StatusCode::BAD_REQUEST,
+            ),
+            ResponseError::Maintenance => error(
+                String::from("Server is in maintenance, please try again later"),
+                StatusCode::SERVICE_UNAVAILABLE,
+            ),
+        }
+    }
+}
+
+#[derive(Serialize, Deserialize)]
+struct ErrorMessage {
+    message: String,
+}
+
+fn error(message: String, status: StatusCode) -> Response {
+    let message = ErrorMessage { message };
+    tide::response::json(message)
+        .with_status(status)
+        .into_response()
+}

meilisearch-http/src/helpers/meilisearch.rs

@@ -0,0 +1,571 @@
+use crate::routes::setting::{RankingOrdering, SettingBody};
+use indexmap::IndexMap;
+use log::error;
+use meilisearch_core::criterion::*;
+use meilisearch_core::Highlight;
+use meilisearch_core::{Index, RankedMap};
+use meilisearch_core::MainT;
+use meilisearch_schema::{Schema, SchemaAttr};
+use serde::{Deserialize, Serialize};
+use serde_json::Value;
+use std::cmp::Ordering;
+use std::collections::{HashMap, HashSet};
+use std::convert::From;
+use std::error;
+use std::fmt;
+use std::time::{Duration, Instant};
+
+#[derive(Debug)]
+pub enum Error {
+    SearchDocuments(String),
+    RetrieveDocument(u64, String),
+    DocumentNotFound(u64),
+    CropFieldWrongType(String),
+    AttributeNotFoundOnDocument(String),
+    AttributeNotFoundOnSchema(String),
+    MissingFilterValue,
+    UnknownFilteredAttribute,
+    Internal(String),
+}
+
+impl error::Error for Error {}
+
+impl fmt::Display for Error {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        use Error::*;
+
+        match self {
+            SearchDocuments(err) => write!(f, "impossible to search documents; {}", err),
+            RetrieveDocument(id, err) => write!(
+                f,
+                "impossible to retrieve the document with id: {}; {}",
+                id, err
+            ),
+            DocumentNotFound(id) => write!(f, "document {} not found", id),
+            CropFieldWrongType(field) => {
+                write!(f, "the field {} cannot be cropped it's not a string", field)
+            }
+            AttributeNotFoundOnDocument(field) => {
+                write!(f, "field {} is not found on document", field)
+            }
+            AttributeNotFoundOnSchema(field) => write!(f, "field {} is not found on schema", field),
+            MissingFilterValue => f.write_str("a filter doesn't have a value to compare it with"),
+            UnknownFilteredAttribute => {
+                f.write_str("a filter is specifying an unknown schema attribute")
+            }
+            Internal(err) => write!(f, "internal error; {}", err),
+        }
+    }
+}
+
+impl From<meilisearch_core::Error> for Error {
+    fn from(error: meilisearch_core::Error) -> Self {
+        Error::Internal(error.to_string())
+    }
+}
+
+pub trait IndexSearchExt {
+    fn new_search(&self, query: String) -> SearchBuilder;
+}
+
+impl IndexSearchExt for Index {
+    fn new_search(&self, query: String) -> SearchBuilder {
+        SearchBuilder {
+            index: self,
+            query,
+            offset: 0,
+            limit: 20,
+            attributes_to_crop: None,
+            attributes_to_retrieve: None,
+            attributes_to_search_in: None,
+            attributes_to_highlight: None,
+            filters: None,
+            timeout: Duration::from_millis(30),
+            matches: false,
+        }
+    }
+}
+
+pub struct SearchBuilder<'a> {
+    index: &'a Index,
+    query: String,
+    offset: usize,
+    limit: usize,
+    attributes_to_crop: Option<HashMap<String, usize>>,
+    attributes_to_retrieve: Option<HashSet<String>>,
+    attributes_to_search_in: Option<HashSet<String>>,
+    attributes_to_highlight: Option<HashSet<String>>,
+    filters: Option<String>,
+    timeout: Duration,
+    matches: bool,
+}
+
+impl<'a> SearchBuilder<'a> {
+    pub fn offset(&mut self, value: usize) -> &SearchBuilder {
+        self.offset = value;
+        self
+    }
+
+    pub fn limit(&mut self, value: usize) -> &SearchBuilder {
+        self.limit = value;
+        self
+    }
+
+    pub fn attributes_to_crop(&mut self, value: HashMap<String, usize>) -> &SearchBuilder {
+        self.attributes_to_crop = Some(value);
+        self
+    }
+
+    pub fn attributes_to_retrieve(&mut self, value: HashSet<String>) -> &SearchBuilder {
+        self.attributes_to_retrieve = Some(value);
+        self
+    }
+
+    pub fn add_retrievable_field(&mut self, value: String) -> &SearchBuilder {
+        let attributes_to_retrieve = self.attributes_to_retrieve.get_or_insert(HashSet::new());
+        attributes_to_retrieve.insert(value);
+        self
+    }
+
+    pub fn attributes_to_search_in(&mut self, value: HashSet<String>) -> &SearchBuilder {
+        self.attributes_to_search_in = Some(value);
+        self
+    }
+
+    pub fn add_attribute_to_search_in(&mut self, value: String) -> &SearchBuilder {
+        let attributes_to_search_in = self.attributes_to_search_in.get_or_insert(HashSet::new());
+        attributes_to_search_in.insert(value);
+        self
+    }
+
+    pub fn attributes_to_highlight(&mut self, value: HashSet<String>) -> &SearchBuilder {
+        self.attributes_to_highlight = Some(value);
+        self
+    }
+
+    pub fn filters(&mut self, value: String) -> &SearchBuilder {
+        self.filters = Some(value);
+        self
+    }
+
+    pub fn timeout(&mut self, value: Duration) -> &SearchBuilder {
+        self.timeout = value;
+        self
+    }
+
+    pub fn get_matches(&mut self) -> &SearchBuilder {
+        self.matches = true;
+        self
+    }
+
+    pub fn search(&self, reader: &heed::RoTxn<MainT>) -> Result<SearchResult, Error> {
+        let schema = self.index.main.schema(reader);
+        let schema = schema.map_err(|e| Error::Internal(e.to_string()))?;
+        let schema = match schema {
+            Some(schema) => schema,
+            None => return Err(Error::Internal(String::from("missing schema"))),
+        };
+
+        let ranked_map = self.index.main.ranked_map(reader);
+        let ranked_map = ranked_map.map_err(|e| Error::Internal(e.to_string()))?;
+        let ranked_map = ranked_map.unwrap_or_default();
+
+        let start = Instant::now();
+
+        // Change criteria
+        let mut query_builder = match self.get_criteria(reader, &ranked_map, &schema)? {
+            Some(criteria) => self.index.query_builder_with_criteria(criteria),
+            None => self.index.query_builder(),
+        };
+
+        // Filter searchable fields
+        if let Some(fields) = &self.attributes_to_search_in {
+            for attribute in fields.iter().filter_map(|f| schema.attribute(f)) {
+                query_builder.add_searchable_attribute(attribute.0);
+            }
+        }
+
+        if let Some(filters) = &self.filters {
+            let mut split = filters.split(':');
+            match (split.next(), split.next()) {
+                (Some(_), None) | (Some(_), Some("")) => return Err(Error::MissingFilterValue),
+                (Some(attr), Some(value)) => {
+                    let ref_reader = reader;
+                    let ref_index = &self.index;
+                    let value = value.trim().to_lowercase();
+
+                    let attr = match schema.attribute(attr) {
+                        Some(attr) => attr,
+                        None => return Err(Error::UnknownFilteredAttribute),
+                    };
+
+                    query_builder.with_filter(move |id| {
+                        let attr = attr;
+                        let index = ref_index;
+                        let reader = ref_reader;
+
+                        match index.document_attribute::<Value>(reader, id, attr) {
+                            Ok(Some(Value::String(s))) => s.to_lowercase() == value,
+                            Ok(Some(Value::Bool(b))) => {
+                                (value == "true" && b) || (value == "false" && !b)
+                            }
+                            Ok(Some(Value::Array(a))) => {
+                                a.into_iter().any(|s| s.as_str() == Some(&value))
+                            }
+                            _ => false,
+                        }
+                    });
+                }
+                (_, _) => (),
+            }
+        }
+
+        query_builder.with_fetch_timeout(self.timeout);
+
+        let docs =
+            query_builder.query(reader, &self.query, self.offset..(self.offset + self.limit));
+
+        let mut hits = Vec::with_capacity(self.limit);
+        for doc in docs.map_err(|e| Error::SearchDocuments(e.to_string()))? {
+            // retrieve the content of document in kv store
+            let mut fields: Option<HashSet<&str>> = None;
+            if let Some(attributes_to_retrieve) = &self.attributes_to_retrieve {
+                let mut set = HashSet::new();
+                for field in attributes_to_retrieve {
+                    set.insert(field.as_str());
+                }
+                fields = Some(set);
+            }
+
+            let document: IndexMap<String, Value> = self
+                .index
+                .document(reader, fields.as_ref(), doc.id)
+                .map_err(|e| Error::RetrieveDocument(doc.id.0, e.to_string()))?
+                .ok_or(Error::DocumentNotFound(doc.id.0))?;
+
+            let mut formatted = document.clone();
+            let mut matches = doc.highlights.clone();
+
+            // Crops fields if needed
+            if let Some(fields) = &self.attributes_to_crop {
+                crop_document(&mut formatted, &mut matches, &schema, fields);
+            }
+
+            // Transform to readable matches
+            let matches = calculate_matches(matches, self.attributes_to_retrieve.clone(), &schema);
+
+            if !self.matches {
+                if let Some(attributes_to_highlight) = &self.attributes_to_highlight {
+                    formatted = calculate_highlights(&formatted, &matches, attributes_to_highlight);
+                }
+            }
+
+            let matches_info = if self.matches { Some(matches) } else { None };
+
+            let hit = SearchHit {
+                document,
+                formatted,
+                matches_info,
+            };
+
+            hits.push(hit);
+        }
+
+        let time_ms = start.elapsed().as_millis() as usize;
+
+        let results = SearchResult {
+            hits,
+            offset: self.offset,
+            limit: self.limit,
+            processing_time_ms: time_ms,
+            query: self.query.to_string(),
+        };
+
+        Ok(results)
+    }
+
+    pub fn get_criteria(
+        &self,
+        reader: &heed::RoTxn<MainT>,
+        ranked_map: &'a RankedMap,
+        schema: &Schema,
+    ) -> Result<Option<Criteria<'a>>, Error> {
+        let current_settings = match self.index.main.customs(reader).unwrap() {
+            Some(bytes) => bincode::deserialize(bytes).unwrap(),
+            None => SettingBody::default(),
+        };
+
+        let ranking_rules = &current_settings.ranking_rules;
+        let ranking_order = &current_settings.ranking_order;
+
+        if let Some(ranking_rules) = ranking_rules {
+            let mut builder = CriteriaBuilder::with_capacity(7 + ranking_rules.len());
+            if let Some(ranking_rules_order) = ranking_order {
+                for rule in ranking_rules_order {
+                    match rule.as_str() {
+                        "_sum_of_typos" => builder.push(SumOfTypos),
+                        "_number_of_words" => builder.push(NumberOfWords),
+                        "_word_proximity" => builder.push(WordsProximity),
+                        "_sum_of_words_attribute" => builder.push(SumOfWordsAttribute),
+                        "_sum_of_words_position" => builder.push(SumOfWordsPosition),
+                        "_exact" => builder.push(Exact),
+                        _ => {
+                            let order = match ranking_rules.get(rule.as_str()) {
+                                Some(o) => o,
+                                None => continue,
+                            };
+
+                            let custom_ranking = match order {
+                                RankingOrdering::Asc => {
+                                    SortByAttr::lower_is_better(&ranked_map, &schema, &rule)
+                                        .unwrap()
+                                }
+                                RankingOrdering::Dsc => {
+                                    SortByAttr::higher_is_better(&ranked_map, &schema, &rule)
+                                        .unwrap()
+                                }
+                            };
+
+                            builder.push(custom_ranking);
+                        }
+                    }
+                }
+                builder.push(DocumentId);
+                return Ok(Some(builder.build()));
+            } else {
+                builder.push(SumOfTypos);
+                builder.push(NumberOfWords);
+                builder.push(WordsProximity);
+                builder.push(SumOfWordsAttribute);
+                builder.push(SumOfWordsPosition);
+                builder.push(Exact);
+                for (rule, order) in ranking_rules.iter() {
+                    let custom_ranking = match order {
+                        RankingOrdering::Asc => {
+                            SortByAttr::lower_is_better(&ranked_map, &schema, &rule).unwrap()
+                        }
+                        RankingOrdering::Dsc => {
+                            SortByAttr::higher_is_better(&ranked_map, &schema, &rule).unwrap()
+                        }
+                    };
+                    builder.push(custom_ranking);
+                }
+                builder.push(DocumentId);
+                return Ok(Some(builder.build()));
+            }
+        }
+
+        Ok(None)
+    }
+}
+
+#[derive(Debug, Clone, Eq, PartialEq, PartialOrd, Serialize, Deserialize)]
+pub struct MatchPosition {
+    pub start: usize,
+    pub length: usize,
+}
+
+impl Ord for MatchPosition {
+    fn cmp(&self, other: &Self) -> Ordering {
+        match self.start.cmp(&other.start) {
+            Ordering::Equal => self.length.cmp(&other.length),
+            _ => self.start.cmp(&other.start),
+        }
+    }
+}
+
+pub type HighlightInfos = HashMap<String, Value>;
+pub type MatchesInfos = HashMap<String, Vec<MatchPosition>>;
+// pub type RankingInfos = HashMap<String, u64>;
+
+#[derive(Debug, Clone, Serialize, Deserialize)]
+pub struct SearchHit {
+    #[serde(flatten)]
+    pub document: IndexMap<String, Value>,
+    #[serde(rename = "_formatted", skip_serializing_if = "IndexMap::is_empty")]
+    pub formatted: IndexMap<String, Value>,
+    #[serde(rename = "_matchesInfo", skip_serializing_if = "Option::is_none")]
+    pub matches_info: Option<MatchesInfos>,
+}
+
+#[derive(Debug, Clone, Serialize, Deserialize)]
+#[serde(rename_all = "camelCase")]
+pub struct SearchResult {
+    pub hits: Vec<SearchHit>,
+    pub offset: usize,
+    pub limit: usize,
+    pub processing_time_ms: usize,
+    pub query: String,
+    // pub parsed_query: String,
+    // pub params: Option<String>,
+}
+
+fn crop_text(
+    text: &str,
+    matches: impl IntoIterator<Item = Highlight>,
+    context: usize,
+) -> (String, Vec<Highlight>) {
+    let mut matches = matches.into_iter().peekable();
+
+    let char_index = matches.peek().map(|m| m.char_index as usize).unwrap_or(0);
+    let start = char_index.saturating_sub(context);
+    let text = text.chars().skip(start).take(context * 2).collect();
+
+    let matches = matches
+        .take_while(|m| (m.char_index as usize) + (m.char_length as usize) <= start + (context * 2))
+        .map(|match_| Highlight {
+            char_index: match_.char_index - start as u16,
+            ..match_
+        })
+        .collect();
+
+    (text, matches)
+}
+
+fn crop_document(
+    document: &mut IndexMap<String, Value>,
+    matches: &mut Vec<Highlight>,
+    schema: &Schema,
+    fields: &HashMap<String, usize>,
+) {
+    matches.sort_unstable_by_key(|m| (m.char_index, m.char_length));
+
+    for (field, length) in fields {
+        let attribute = match schema.attribute(field) {
+            Some(attribute) => attribute,
+            None => continue,
+        };
+
+        let selected_matches = matches
+            .iter()
+            .filter(|m| SchemaAttr::new(m.attribute) == attribute)
+            .cloned();
+
+        if let Some(Value::String(ref mut original_text)) = document.get_mut(field) {
+            let (cropped_text, cropped_matches) =
+                crop_text(original_text, selected_matches, *length);
+
+            *original_text = cropped_text;
+
+            matches.retain(|m| SchemaAttr::new(m.attribute) != attribute);
+            matches.extend_from_slice(&cropped_matches);
+        }
+    }
+}
+
+fn calculate_matches(
+    matches: Vec<Highlight>,
+    attributes_to_retrieve: Option<HashSet<String>>,
+    schema: &Schema,
+) -> MatchesInfos {
+    let mut matches_result: HashMap<String, Vec<MatchPosition>> = HashMap::new();
+    for m in matches.iter() {
+        let attribute = schema
+            .attribute_name(SchemaAttr::new(m.attribute))
+            .to_string();
+        if let Some(attributes_to_retrieve) = attributes_to_retrieve.clone() {
+            if !attributes_to_retrieve.contains(attribute.as_str()) {
+                continue;
+            }
+        };
+        if let Some(pos) = matches_result.get_mut(&attribute) {
+            pos.push(MatchPosition {
+                start: m.char_index as usize,
+                length: m.char_length as usize,
+            });
+        } else {
+            let mut positions = Vec::new();
+            positions.push(MatchPosition {
+                start: m.char_index as usize,
+                length: m.char_length as usize,
+            });
+            matches_result.insert(attribute, positions);
+        }
+    }
+    for (_, val) in matches_result.iter_mut() {
+        val.sort_unstable();
+        val.dedup();
+    }
+    matches_result
+}
+
+fn calculate_highlights(
+    document: &IndexMap<String, Value>,
+    matches: &MatchesInfos,
+    attributes_to_highlight: &HashSet<String>,
+) -> IndexMap<String, Value> {
+    let mut highlight_result = IndexMap::new();
+
+    for (attribute, matches) in matches.iter() {
+        if attributes_to_highlight.contains(attribute) {
+            if let Some(Value::String(value)) = document.get(attribute) {
+                let value: Vec<_> = value.chars().collect();
+                let mut highlighted_value = String::new();
+                let mut index = 0;
+                for m in matches {
+                    if m.start >= index {
+                        let before = value.get(index..m.start);
+                        let highlighted = value.get(m.start..(m.start + m.length));
+                        if let (Some(before), Some(highlighted)) = (before, highlighted) {
+                            highlighted_value.extend(before);
+                            highlighted_value.push_str("<em>");
+                            highlighted_value.extend(highlighted);
+                            highlighted_value.push_str("</em>");
+                            index = m.start + m.length;
+                        } else {
+                            error!("value: {:?}; index: {:?}, match: {:?}", value, index, m);
+                        }
+                    }
+                }
+                highlighted_value.extend(value[index..].iter());
+                highlight_result.insert(attribute.to_string(), Value::String(highlighted_value));
+            };
+        }
+    }
+
+    highlight_result
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn calculate_highlights() {
+        let data = r#"{
+            "title": "Fondation (Isaac ASIMOV)",
+            "description": "En ce début de trentième millénaire, l'Empire n'a jamais été aussi puissant, aussi étendu à travers toute la galaxie. C'est dans sa capitale, Trantor, que l'éminent savant Hari Seldon invente la psychohistoire, une science toute nouvelle, à base de psychologie et de mathématiques, qui lui permet de prédire l'avenir... C'est-à-dire l'effondrement de l'Empire d'ici cinq siècles et au-delà, trente mille années de chaos et de ténèbres. Pour empêcher cette catastrophe et sauver la civilisation, Seldon crée la Fondation."
+        }"#;
+
+        let document: IndexMap<String, Value> = serde_json::from_str(data).unwrap();
+        let mut attributes_to_highlight = HashSet::new();
+        attributes_to_highlight.insert("title".to_string());
+        attributes_to_highlight.insert("description".to_string());
+
+        let mut matches = HashMap::new();
+
+        let mut m = Vec::new();
+        m.push(MatchPosition {
+            start: 0,
+            length: 9,
+        });
+        matches.insert("title".to_string(), m);
+
+        let mut m = Vec::new();
+        m.push(MatchPosition {
+            start: 510,
+            length: 9,
+        });
+        matches.insert("description".to_string(), m);
+        let result = super::calculate_highlights(&document, &matches, &attributes_to_highlight);
+
+        let mut result_expected = IndexMap::new();
+        result_expected.insert(
+            "title".to_string(),
+            Value::String("<em>Fondation</em> (Isaac ASIMOV)".to_string()),
+        );
+        result_expected.insert("description".to_string(), Value::String("En ce début de trentième millénaire, l'Empire n'a jamais été aussi puissant, aussi étendu à travers toute la galaxie. C'est dans sa capitale, Trantor, que l'éminent savant Hari Seldon invente la psychohistoire, une science toute nouvelle, à base de psychologie et de mathématiques, qui lui permet de prédire l'avenir... C'est-à-dire l'effondrement de l'Empire d'ici cinq siècles et au-delà, trente mille années de chaos et de ténèbres. Pour empêcher cette catastrophe et sauver la civilisation, Seldon crée la <em>Fondation</em>.".to_string()));
+
+        assert_eq!(result, result_expected);
+    }
+}

meilisearch-http/src/helpers/mod.rs

@@ -0,0 +1,2 @@
+pub mod meilisearch;
+pub mod tide;

meilisearch-http/src/helpers/tide.rs

@@ -0,0 +1,117 @@
+use crate::error::{ResponseError, SResult};
+use crate::models::token::*;
+use crate::Data;
+use chrono::Utc;
+use heed::types::{SerdeBincode, Str};
+use meilisearch_core::Index;
+use serde_json::Value;
+use tide::Context;
+
+pub trait ContextExt {
+    fn is_allowed(&self, acl: ACL) -> SResult<()>;
+    fn header(&self, name: &str) -> Result<String, ResponseError>;
+    fn url_param(&self, name: &str) -> Result<String, ResponseError>;
+    fn index(&self) -> Result<Index, ResponseError>;
+    fn identifier(&self) -> Result<String, ResponseError>;
+}
+
+impl ContextExt for Context<Data> {
+    fn is_allowed(&self, acl: ACL) -> SResult<()> {
+        let api_key = match &self.state().api_key {
+            Some(api_key) => api_key,
+            None => return Ok(()),
+        };
+
+        let user_api_key = self.header("X-Meili-API-Key")?;
+        if user_api_key == *api_key {
+            return Ok(());
+        }
+        let request_index: Option<String> = None; //self.param::<String>("index").ok();
+
+        let db = &self.state().db;
+        let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+
+        let token_key = format!("{}{}", TOKEN_PREFIX_KEY, user_api_key);
+
+        let token_config = db
+            .common_store()
+            .get::<_, Str, SerdeBincode<Token>>(&reader, &token_key)
+            .map_err(ResponseError::internal)?
+            .ok_or(ResponseError::invalid_token(format!(
+                "Api key does not exist: {}",
+                user_api_key
+            )))?;
+
+        if token_config.revoked {
+            return Err(ResponseError::invalid_token("token revoked"));
+        }
+
+        if let Some(index) = request_index {
+            if !token_config
+                .indexes
+                .iter()
+                .any(|r| match_wildcard(&r, &index))
+            {
+                return Err(ResponseError::invalid_token(
+                    "token is not allowed to access to this index",
+                ));
+            }
+        }
+
+        if token_config.expires_at < Utc::now() {
+            return Err(ResponseError::invalid_token("token expired"));
+        }
+
+        if token_config.acl.contains(&ACL::All) {
+            return Ok(());
+        }
+
+        if !token_config.acl.contains(&acl) {
+            return Err(ResponseError::invalid_token("token do not have this ACL"));
+        }
+
+        Ok(())
+    }
+
+    fn header(&self, name: &str) -> Result<String, ResponseError> {
+        let header = self
+            .headers()
+            .get(name)
+            .ok_or(ResponseError::missing_header(name))?
+            .to_str()
+            .map_err(|_| ResponseError::missing_header("X-Meili-API-Key"))?
+            .to_string();
+        Ok(header)
+    }
+
+    fn url_param(&self, name: &str) -> Result<String, ResponseError> {
+        let param = self
+            .param::<String>(name)
+            .map_err(|e| ResponseError::bad_parameter(name, e))?;
+        Ok(param)
+    }
+
+    fn index(&self) -> Result<Index, ResponseError> {
+        let index_uid = self.url_param("index")?;
+        let index = self
+            .state()
+            .db
+            .open_index(&index_uid)
+            .ok_or(ResponseError::index_not_found(index_uid))?;
+        Ok(index)
+    }
+
+    fn identifier(&self) -> Result<String, ResponseError> {
+        let name = self
+            .param::<Value>("identifier")
+            .as_ref()
+            .map(meilisearch_core::serde::value_to_string)
+            .map_err(|e| ResponseError::bad_parameter("identifier", e))?
+            .ok_or(ResponseError::bad_parameter(
+                "identifier",
+                "missing parameter",
+            ))?;
+
+        Ok(name)
+    }
+}

meilisearch-http/src/lib.rs

@@ -0,0 +1,8 @@
+pub mod data;
+pub mod error;
+pub mod helpers;
+pub mod models;
+pub mod option;
+pub mod routes;
+
+pub use self::data::Data;

meilisearch-http/src/main.rs

@@ -0,0 +1,54 @@
+use std::env::VarError::NotPresent;
+use std::{env, thread};
+
+use http::header::HeaderValue;
+use log::info;
+use main_error::MainError;
+use structopt::StructOpt;
+use tide::middleware::{CorsMiddleware, CorsOrigin};
+use tide_log::RequestLogger;
+
+use meilisearch_http::data::Data;
+use meilisearch_http::option::Opt;
+use meilisearch_http::routes;
+use meilisearch_http::routes::index::index_update_callback;
+
+mod analytics;
+
+#[cfg(target_os = "linux")]
+#[global_allocator]
+static ALLOC: jemallocator::Jemalloc = jemallocator::Jemalloc;
+
+pub fn main() -> Result<(), MainError> {
+    env_logger::init();
+
+    let opt = Opt::from_args();
+    let data = Data::new(opt.clone());
+
+    if env::var("MEILI_NO_ANALYTICS") == Err(NotPresent) {
+        thread::spawn(|| analytics::analytics_sender());
+    }
+
+    let data_cloned = data.clone();
+    data.db.set_update_callback(Box::new(move |name, status| {
+        index_update_callback(name, &data_cloned, status);
+    }));
+
+    let mut app = tide::App::with_state(data);
+
+    app.middleware(
+        CorsMiddleware::new()
+            .allow_origin(CorsOrigin::from("*"))
+            .allow_methods(HeaderValue::from_static("GET, POST, OPTIONS")),
+    );
+    app.middleware(RequestLogger::new());
+    app.middleware(tide_compression::Compression::new());
+    app.middleware(tide_compression::Decompression::new());
+
+    routes::load_routes(&mut app);
+
+    info!("Server HTTP enabled");
+    app.run(opt.http_addr)?;
+
+    Ok(())
+}

meilisearch-http/src/models/mod.rs

@@ -0,0 +1,3 @@
+pub mod schema;
+pub mod token;
+pub mod update_operation;

meilisearch-http/src/models/schema.rs

@@ -0,0 +1,118 @@
+use std::collections::HashSet;
+
+use indexmap::IndexMap;
+use meilisearch_schema::{Schema, SchemaBuilder, SchemaProps};
+use serde::{Deserialize, Serialize};
+
+#[derive(Debug, Clone, Eq, PartialEq, Hash, Serialize, Deserialize)]
+#[serde(rename_all = "camelCase")]
+pub enum FieldProperties {
+    Identifier,
+    Indexed,
+    Displayed,
+    Ranked,
+}
+
+#[derive(Debug, Clone, Eq, PartialEq, Serialize, Deserialize)]
+pub struct SchemaBody(IndexMap<String, HashSet<FieldProperties>>);
+
+impl From<Schema> for SchemaBody {
+    fn from(value: Schema) -> SchemaBody {
+        let mut map = IndexMap::new();
+        for (name, _attr, props) in value.iter() {
+            let old_properties = map.entry(name.to_owned()).or_insert(HashSet::new());
+            if props.is_indexed() {
+                old_properties.insert(FieldProperties::Indexed);
+            }
+            if props.is_displayed() {
+                old_properties.insert(FieldProperties::Displayed);
+            }
+            if props.is_ranked() {
+                old_properties.insert(FieldProperties::Ranked);
+            }
+        }
+        let old_properties = map
+            .entry(value.identifier_name().to_string())
+            .or_insert(HashSet::new());
+        old_properties.insert(FieldProperties::Identifier);
+        old_properties.insert(FieldProperties::Displayed);
+        SchemaBody(map)
+    }
+}
+
+impl Into<Schema> for SchemaBody {
+    fn into(self) -> Schema {
+        let mut identifier = "documentId".to_string();
+        let mut attributes = IndexMap::new();
+        for (field, properties) in self.0 {
+            let mut indexed = false;
+            let mut displayed = false;
+            let mut ranked = false;
+            for property in properties {
+                match property {
+                    FieldProperties::Indexed => indexed = true,
+                    FieldProperties::Displayed => displayed = true,
+                    FieldProperties::Ranked => ranked = true,
+                    FieldProperties::Identifier => identifier = field.clone(),
+                }
+            }
+            attributes.insert(
+                field,
+                SchemaProps {
+                    indexed,
+                    displayed,
+                    ranked,
+                },
+            );
+        }
+
+        let mut builder = SchemaBuilder::with_identifier(identifier);
+        for (field, props) in attributes {
+            builder.new_attribute(field, props);
+        }
+        builder.build()
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_schema_body_conversion() {
+        let schema_body = r#"
+        {
+            "id": ["identifier", "indexed", "displayed"],
+            "title": ["indexed", "displayed"],
+            "date": ["displayed"]
+        }
+        "#;
+
+        let schema_builder = r#"
+        {
+            "identifier": "id",
+            "attributes": {
+                "id": {
+                    "indexed": true,
+                    "displayed": true
+                },
+                "title": {
+                    "indexed": true,
+                    "displayed": true
+                },
+                "date": {
+                    "displayed": true
+                }
+            }
+        }
+        "#;
+
+        let schema_body: SchemaBody = serde_json::from_str(schema_body).unwrap();
+        let schema_builder: SchemaBuilder = serde_json::from_str(schema_builder).unwrap();
+
+        let schema_from_body: Schema = schema_body.into();
+        let schema_from_builder: Schema = schema_builder.build();
+
+        assert_eq!(schema_from_body, schema_from_builder);
+    }
+}

meilisearch-http/src/models/token.rs

@@ -0,0 +1,72 @@
+use chrono::{DateTime, Utc};
+use serde::{Deserialize, Serialize};
+
+pub const TOKEN_PREFIX_KEY: &str = "_token_";
+
+#[derive(Debug, Clone, Eq, PartialEq, Serialize, Deserialize)]
+#[serde(rename_all = "camelCase")]
+pub enum ACL {
+    IndexesRead,
+    IndexesWrite,
+    DocumentsRead,
+    DocumentsWrite,
+    SettingsRead,
+    SettingsWrite,
+    Admin,
+    #[serde(rename = "*")]
+    All,
+}
+
+pub type Wildcard = String;
+
+#[derive(Debug, Clone, Serialize, Deserialize)]
+#[serde(rename_all = "camelCase")]
+pub struct Token {
+    pub key: String,
+    pub description: String,
+    pub acl: Vec<ACL>,
+    pub indexes: Vec<Wildcard>,
+    pub created_at: DateTime<Utc>,
+    pub updated_at: DateTime<Utc>,
+    pub expires_at: DateTime<Utc>,
+    pub revoked: bool,
+}
+
+fn cleanup_wildcard(input: &str) -> (bool, &str, bool) {
+    let first = input.chars().next().filter(|&c| c == '*').is_some();
+    let last = input.chars().last().filter(|&c| c == '*').is_some();
+    let bound_last = std::cmp::max(input.len().saturating_sub(last as usize), first as usize);
+    let output = input.get(first as usize..bound_last).unwrap();
+    (first, output, last)
+}
+
+pub fn match_wildcard(pattern: &str, input: &str) -> bool {
+    let (first, pattern, last) = cleanup_wildcard(pattern);
+
+    match (first, last) {
+        (false, false) => pattern == input,
+        (true, false) => input.ends_with(pattern),
+        (false, true) => input.starts_with(pattern),
+        (true, true) => input.contains(pattern),
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_match_wildcard() {
+        assert!(match_wildcard("*", "qqq"));
+        assert!(match_wildcard("*", ""));
+        assert!(match_wildcard("*ab", "qqqab"));
+        assert!(match_wildcard("*ab*", "qqqabqq"));
+        assert!(match_wildcard("ab*", "abqqq"));
+        assert!(match_wildcard("**", "ab"));
+        assert!(match_wildcard("ab", "ab"));
+        assert!(match_wildcard("ab*", "ab"));
+        assert!(match_wildcard("*ab", "ab"));
+        assert!(match_wildcard("*ab*", "ab"));
+        assert!(match_wildcard("*😆*", "ab😆dsa"));
+    }
+}

meilisearch-http/src/models/update_operation.rs

@@ -0,0 +1,33 @@
+use std::fmt;
+
+#[allow(dead_code)]
+#[derive(Debug)]
+pub enum UpdateOperation {
+    ClearAllDocuments,
+    DocumentsAddition,
+    DocumentsDeletion,
+    SynonymsAddition,
+    SynonymsDeletion,
+    StopWordsAddition,
+    StopWordsDeletion,
+    Schema,
+    Config,
+}
+
+impl fmt::Display for UpdateOperation {
+    fn fmt(&self, f: &mut fmt::Formatter) -> std::fmt::Result {
+        use UpdateOperation::*;
+
+        match self {
+            ClearAllDocuments => write!(f, "ClearAllDocuments"),
+            DocumentsAddition => write!(f, "DocumentsAddition"),
+            DocumentsDeletion => write!(f, "DocumentsDeletion"),
+            SynonymsAddition => write!(f, "SynonymsAddition"),
+            SynonymsDeletion => write!(f, "SynonymsDelettion"),
+            StopWordsAddition => write!(f, "StopWordsAddition"),
+            StopWordsDeletion => write!(f, "StopWordsDeletion"),
+            Schema => write!(f, "Schema"),
+            Config => write!(f, "Config"),
+        }
+    }
+}

meilisearch-http/src/option.rs

@@ -0,0 +1,20 @@
+use structopt::StructOpt;
+
+#[derive(Debug, Clone, StructOpt)]
+pub struct Opt {
+    /// The destination where the database must be created.
+    #[structopt(long, env = "MEILI_DB_PATH", default_value = "/tmp/meilisearch")]
+    pub db_path: String,
+
+    /// The address on which the http server will listen.
+    #[structopt(long, env = "MEILI_HTTP_ADDR", default_value = "127.0.0.1:8080")]
+    pub http_addr: String,
+
+    /// The master key allowing you to do everything on the server.
+    #[structopt(long, env = "MEILI_API_KEY")]
+    pub api_key: Option<String>,
+
+    /// Do not send analytics to Meili.
+    #[structopt(long, env = "MEILI_NO_ANALYTICS")]
+    pub no_analytics: bool,
+}

meilisearch-http/src/routes/document.rs

@@ -0,0 +1,245 @@
+use std::collections::{BTreeSet, HashSet};
+
+use http::StatusCode;
+use indexmap::IndexMap;
+use serde::{Deserialize, Serialize};
+use serde_json::Value;
+use tide::querystring::ContextExt as QSContextExt;
+use tide::response::IntoResponse;
+use tide::{Context, Response};
+
+use crate::error::{ResponseError, SResult};
+use crate::helpers::tide::ContextExt;
+use crate::models::token::ACL::*;
+use crate::Data;
+
+pub async fn get_document(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(DocumentsRead)?;
+
+    let index = ctx.index()?;
+
+    let identifier = ctx.identifier()?;
+    let document_id = meilisearch_core::serde::compute_document_id(identifier.clone());
+
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+
+    let response = index
+        .document::<IndexMap<String, Value>>(&reader, None, document_id)
+        .map_err(ResponseError::internal)?
+        .ok_or(ResponseError::document_not_found(&identifier))?;
+
+    if response.is_empty() {
+        return Err(ResponseError::document_not_found(identifier));
+    }
+
+    Ok(tide::response::json(response))
+}
+
+#[derive(Default, Serialize)]
+#[serde(rename_all = "camelCase")]
+pub struct IndexUpdateResponse {
+    pub update_id: u64,
+}
+
+pub async fn delete_document(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(DocumentsWrite)?;
+
+    let index = ctx.index()?;
+    let identifier = ctx.identifier()?;
+    let document_id = meilisearch_core::serde::compute_document_id(identifier.clone());
+
+    let db = &ctx.state().db;
+    let mut update_writer = db.update_write_txn().map_err(ResponseError::internal)?;
+
+    let mut documents_deletion = index.documents_deletion();
+    documents_deletion.delete_document_by_id(document_id);
+    let update_id = documents_deletion
+        .finalize(&mut update_writer)
+        .map_err(ResponseError::internal)?;
+
+    update_writer.commit().map_err(ResponseError::internal)?;
+
+    let response_body = IndexUpdateResponse { update_id };
+    Ok(tide::response::json(response_body)
+        .with_status(StatusCode::ACCEPTED)
+        .into_response())
+}
+
+#[derive(Default, Deserialize)]
+#[serde(rename_all = "camelCase", deny_unknown_fields)]
+struct BrowseQuery {
+    offset: Option<usize>,
+    limit: Option<usize>,
+    attributes_to_retrieve: Option<String>,
+}
+
+pub async fn get_all_documents(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(DocumentsRead)?;
+
+    let index = ctx.index()?;
+    let query: BrowseQuery = ctx.url_query().unwrap_or(BrowseQuery::default());
+
+    let offset = query.offset.unwrap_or(0);
+    let limit = query.limit.unwrap_or(20);
+
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+
+    let documents_ids: Result<BTreeSet<_>, _> =
+        match index.documents_fields_counts.documents_ids(&reader) {
+            Ok(documents_ids) => documents_ids.skip(offset).take(limit).collect(),
+            Err(e) => return Err(ResponseError::internal(e)),
+        };
+
+    let documents_ids = match documents_ids {
+        Ok(documents_ids) => documents_ids,
+        Err(e) => return Err(ResponseError::internal(e)),
+    };
+
+    let mut response_body = Vec::<IndexMap<String, Value>>::new();
+
+    if let Some(attributes) = query.attributes_to_retrieve {
+        let attributes = attributes.split(',').collect::<HashSet<&str>>();
+        for document_id in documents_ids {
+            if let Ok(Some(document)) = index.document(&reader, Some(&attributes), document_id) {
+                response_body.push(document);
+            }
+        }
+    } else {
+        for document_id in documents_ids {
+            if let Ok(Some(document)) = index.document(&reader, None, document_id) {
+                response_body.push(document);
+            }
+        }
+    }
+
+    Ok(tide::response::json(response_body))
+}
+
+fn infered_schema(document: &IndexMap<String, Value>) -> Option<meilisearch_schema::Schema> {
+    use meilisearch_schema::{SchemaBuilder, DISPLAYED, INDEXED};
+
+    let mut identifier = None;
+    for key in document.keys() {
+        if identifier.is_none() && key.to_lowercase().contains("id") {
+            identifier = Some(key);
+        }
+    }
+
+    match identifier {
+        Some(identifier) => {
+            let mut builder = SchemaBuilder::with_identifier(identifier);
+            for key in document.keys() {
+                builder.new_attribute(key, DISPLAYED | INDEXED);
+            }
+            Some(builder.build())
+        }
+        None => None,
+    }
+}
+
+async fn update_multiple_documents(mut ctx: Context<Data>, is_partial: bool) -> SResult<Response> {
+    ctx.is_allowed(DocumentsWrite)?;
+
+    let data: Vec<IndexMap<String, Value>> =
+        ctx.body_json().await.map_err(ResponseError::bad_request)?;
+    let index = ctx.index()?;
+
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+    let mut update_writer = db.update_write_txn().map_err(ResponseError::internal)?;
+
+    let current_schema = index
+        .main
+        .schema(&reader)
+        .map_err(ResponseError::internal)?;
+    if current_schema.is_none() {
+        match data.first().and_then(infered_schema) {
+            Some(schema) => {
+                index
+                    .schema_update(&mut update_writer, schema)
+                    .map_err(ResponseError::internal)?;
+            }
+            None => return Err(ResponseError::bad_request("Could not infer a schema")),
+        }
+    }
+
+    let mut document_addition = if is_partial {
+        index.documents_partial_addition()
+    } else {
+        index.documents_addition()
+    };
+
+    for document in data {
+        document_addition.update_document(document);
+    }
+
+    let update_id = document_addition
+        .finalize(&mut update_writer)
+        .map_err(ResponseError::internal)?;
+
+    update_writer.commit().map_err(ResponseError::internal)?;
+
+    let response_body = IndexUpdateResponse { update_id };
+    Ok(tide::response::json(response_body)
+        .with_status(StatusCode::ACCEPTED)
+        .into_response())
+}
+
+pub async fn add_or_replace_multiple_documents(ctx: Context<Data>) -> SResult<Response> {
+    update_multiple_documents(ctx, false).await
+}
+
+pub async fn add_or_update_multiple_documents(ctx: Context<Data>) -> SResult<Response> {
+    update_multiple_documents(ctx, true).await
+}
+
+pub async fn delete_multiple_documents(mut ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(DocumentsWrite)?;
+
+    let data: Vec<Value> = ctx.body_json().await.map_err(ResponseError::bad_request)?;
+    let index = ctx.index()?;
+
+    let db = &ctx.state().db;
+    let mut writer = db.update_write_txn().map_err(ResponseError::internal)?;
+
+    let mut documents_deletion = index.documents_deletion();
+
+    for identifier in data {
+        if let Some(identifier) = meilisearch_core::serde::value_to_string(&identifier) {
+            documents_deletion
+                .delete_document_by_id(meilisearch_core::serde::compute_document_id(identifier));
+        }
+    }
+
+    let update_id = documents_deletion
+        .finalize(&mut writer)
+        .map_err(ResponseError::internal)?;
+
+    writer.commit().map_err(ResponseError::internal)?;
+
+    let response_body = IndexUpdateResponse { update_id };
+    Ok(tide::response::json(response_body)
+        .with_status(StatusCode::ACCEPTED)
+        .into_response())
+}
+
+pub async fn clear_all_documents(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(DocumentsWrite)?;
+
+    let index = ctx.index()?;
+
+    let db = &ctx.state().db;
+    let mut writer = db.update_write_txn().map_err(ResponseError::internal)?;
+
+    let update_id = index
+        .clear_all(&mut writer)
+        .map_err(ResponseError::internal)?;
+    writer.commit().map_err(ResponseError::internal)?;
+
+    let response_body = IndexUpdateResponse { update_id };
+    Ok(tide::response::json(response_body)
+        .with_status(StatusCode::ACCEPTED)
+        .into_response())
+}

meilisearch-http/src/routes/health.rs

@@ -0,0 +1,76 @@
+use crate::error::{ResponseError, SResult};
+use crate::helpers::tide::ContextExt;
+use crate::models::token::ACL::*;
+use crate::Data;
+
+use heed::types::{Str, Unit};
+use serde::Deserialize;
+use tide::Context;
+
+const UNHEALTHY_KEY: &str = "_is_unhealthy";
+
+pub async fn get_health(ctx: Context<Data>) -> SResult<()> {
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+
+    let common_store = ctx.state().db.common_store();
+
+    if let Ok(Some(_)) = common_store.get::<_, Str, Unit>(&reader, UNHEALTHY_KEY) {
+        return Err(ResponseError::Maintenance);
+    }
+
+    Ok(())
+}
+
+pub async fn set_healthy(ctx: Context<Data>) -> SResult<()> {
+    ctx.is_allowed(Admin)?;
+
+    let db = &ctx.state().db;
+    let mut writer = db.main_write_txn().map_err(ResponseError::internal)?;
+
+    let common_store = ctx.state().db.common_store();
+    match common_store.delete::<_, Str>(&mut writer, UNHEALTHY_KEY) {
+        Ok(_) => (),
+        Err(e) => return Err(ResponseError::internal(e)),
+    }
+
+    if let Err(e) = writer.commit() {
+        return Err(ResponseError::internal(e));
+    }
+
+    Ok(())
+}
+
+pub async fn set_unhealthy(ctx: Context<Data>) -> SResult<()> {
+    ctx.is_allowed(Admin)?;
+
+    let db = &ctx.state().db;
+    let mut writer = db.main_write_txn().map_err(ResponseError::internal)?;
+
+    let common_store = ctx.state().db.common_store();
+
+    if let Err(e) = common_store.put::<_, Str, Unit>(&mut writer, UNHEALTHY_KEY, &()) {
+        return Err(ResponseError::internal(e));
+    }
+
+    if let Err(e) = writer.commit() {
+        return Err(ResponseError::internal(e));
+    }
+
+    Ok(())
+}
+
+#[derive(Deserialize, Clone)]
+struct HealtBody {
+    health: bool,
+}
+
+pub async fn change_healthyness(mut ctx: Context<Data>) -> SResult<()> {
+    let body: HealtBody = ctx.body_json().await.map_err(ResponseError::bad_request)?;
+
+    if body.health {
+        set_healthy(ctx).await
+    } else {
+        set_unhealthy(ctx).await
+    }
+}

meilisearch-http/src/routes/index.rs

@@ -0,0 +1,439 @@
+use chrono::{DateTime, Utc};
+use http::StatusCode;
+use log::error;
+use meilisearch_core::ProcessedUpdateResult;
+use meilisearch_schema::{Schema, SchemaBuilder};
+use rand::seq::SliceRandom;
+use serde::{Deserialize, Serialize};
+use serde_json::json;
+use tide::querystring::ContextExt as QSContextExt;
+use tide::response::IntoResponse;
+use tide::{Context, Response};
+
+use crate::error::{ResponseError, SResult};
+use crate::helpers::tide::ContextExt;
+use crate::models::schema::SchemaBody;
+use crate::models::token::ACL::*;
+use crate::routes::document::IndexUpdateResponse;
+use crate::Data;
+
+fn generate_uid() -> String {
+    let mut rng = rand::thread_rng();
+    let sample = b"abcdefghijklmnopqrstuvwxyz0123456789";
+    sample
+        .choose_multiple(&mut rng, 8)
+        .map(|c| *c as char)
+        .collect()
+}
+
+pub async fn list_indexes(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(IndexesRead)?;
+
+    let indexes_uids = ctx.state().db.indexes_uids();
+
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+
+    let mut response_body = Vec::new();
+
+    for index_uid in indexes_uids {
+        let index = ctx.state().db.open_index(&index_uid);
+
+        match index {
+            Some(index) => {
+                let name = index
+                    .main
+                    .name(&reader)
+                    .map_err(ResponseError::internal)?
+                    .ok_or(ResponseError::internal("'name' not found"))?;
+                let created_at = index
+                    .main
+                    .created_at(&reader)
+                    .map_err(ResponseError::internal)?
+                    .ok_or(ResponseError::internal("'created_at' date not found"))?;
+                let updated_at = index
+                    .main
+                    .updated_at(&reader)
+                    .map_err(ResponseError::internal)?
+                    .ok_or(ResponseError::internal("'updated_at' date not found"))?;
+
+                let index_reponse = IndexResponse {
+                    name,
+                    uid: index_uid,
+                    created_at,
+                    updated_at,
+                };
+                response_body.push(index_reponse);
+            }
+            None => error!(
+                "Index {} is referenced in the indexes list but cannot be found",
+                index_uid
+            ),
+        }
+    }
+
+    Ok(tide::response::json(response_body))
+}
+
+#[derive(Debug, Serialize)]
+#[serde(rename_all = "camelCase")]
+struct IndexResponse {
+    name: String,
+    uid: String,
+    created_at: DateTime<Utc>,
+    updated_at: DateTime<Utc>,
+}
+
+pub async fn get_index(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(IndexesRead)?;
+
+    let index = ctx.index()?;
+
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+
+    let uid = ctx.url_param("index")?;
+    let name = index
+        .main
+        .name(&reader)
+        .map_err(ResponseError::internal)?
+        .ok_or(ResponseError::internal("'name' not found"))?;
+    let created_at = index
+        .main
+        .created_at(&reader)
+        .map_err(ResponseError::internal)?
+        .ok_or(ResponseError::internal("'created_at' date not found"))?;
+    let updated_at = index
+        .main
+        .updated_at(&reader)
+        .map_err(ResponseError::internal)?
+        .ok_or(ResponseError::internal("'updated_at' date not found"))?;
+
+    let response_body = IndexResponse {
+        name,
+        uid,
+        created_at,
+        updated_at,
+    };
+
+    Ok(tide::response::json(response_body))
+}
+
+#[derive(Debug, Deserialize)]
+#[serde(rename_all = "camelCase", deny_unknown_fields)]
+struct IndexCreateRequest {
+    name: String,
+    uid: Option<String>,
+    schema: Option<SchemaBody>,
+}
+
+#[derive(Debug, Serialize)]
+#[serde(rename_all = "camelCase")]
+struct IndexCreateResponse {
+    name: String,
+    uid: String,
+    schema: Option<SchemaBody>,
+    #[serde(skip_serializing_if = "Option::is_none")]
+    update_id: Option<u64>,
+    created_at: DateTime<Utc>,
+    updated_at: DateTime<Utc>,
+}
+
+pub async fn create_index(mut ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(IndexesWrite)?;
+
+    let body = ctx
+        .body_json::<IndexCreateRequest>()
+        .await
+        .map_err(ResponseError::bad_request)?;
+
+    let db = &ctx.state().db;
+
+    let uid = match body.uid {
+        Some(uid) => uid,
+        None => loop {
+            let uid = generate_uid();
+            if db.open_index(&uid).is_none() {
+                break uid;
+            }
+        },
+    };
+
+    let created_index = match db.create_index(&uid) {
+        Ok(index) => index,
+        Err(e) => return Err(ResponseError::create_index(e)),
+    };
+
+    let mut writer = db.main_write_txn().map_err(ResponseError::internal)?;
+    let mut update_writer = db.update_write_txn().map_err(ResponseError::internal)?;
+
+    created_index
+        .main
+        .put_name(&mut writer, &body.name)
+        .map_err(ResponseError::internal)?;
+    created_index
+        .main
+        .put_created_at(&mut writer)
+        .map_err(ResponseError::internal)?;
+    created_index
+        .main
+        .put_updated_at(&mut writer)
+        .map_err(ResponseError::internal)?;
+
+    let schema: Option<Schema> = body.schema.clone().map(Into::into);
+    let mut response_update_id = None;
+    if let Some(schema) = schema {
+        let update_id = created_index
+            .schema_update(&mut update_writer, schema)
+            .map_err(ResponseError::internal)?;
+        response_update_id = Some(update_id)
+    }
+
+    writer.commit().map_err(ResponseError::internal)?;
+    update_writer.commit().map_err(ResponseError::internal)?;
+
+    let response_body = IndexCreateResponse {
+        name: body.name,
+        uid,
+        schema: body.schema,
+        update_id: response_update_id,
+        created_at: Utc::now(),
+        updated_at: Utc::now(),
+    };
+
+    Ok(tide::response::json(response_body)
+        .with_status(StatusCode::CREATED)
+        .into_response())
+}
+
+#[derive(Debug, Deserialize)]
+#[serde(rename_all = "camelCase", deny_unknown_fields)]
+struct UpdateIndexRequest {
+    name: String,
+}
+
+#[derive(Debug, Serialize)]
+#[serde(rename_all = "camelCase")]
+struct UpdateIndexResponse {
+    name: String,
+    uid: String,
+    created_at: DateTime<Utc>,
+    updated_at: DateTime<Utc>,
+}
+
+pub async fn update_index(mut ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(IndexesWrite)?;
+
+    let body = ctx
+        .body_json::<UpdateIndexRequest>()
+        .await
+        .map_err(ResponseError::bad_request)?;
+
+    let index_uid = ctx.url_param("index")?;
+    let index = ctx.index()?;
+
+    let db = &ctx.state().db;
+    let mut writer = db.main_write_txn().map_err(ResponseError::internal)?;
+
+    index
+        .main
+        .put_name(&mut writer, &body.name)
+        .map_err(ResponseError::internal)?;
+
+    index
+        .main
+        .put_updated_at(&mut writer)
+        .map_err(ResponseError::internal)?;
+
+    writer.commit().map_err(ResponseError::internal)?;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+
+    let created_at = index
+        .main
+        .created_at(&reader)
+        .map_err(ResponseError::internal)?
+        .ok_or(ResponseError::internal("'created_at' date not found"))?;
+    let updated_at = index
+        .main
+        .updated_at(&reader)
+        .map_err(ResponseError::internal)?
+        .ok_or(ResponseError::internal("'updated_at' date not found"))?;
+
+    let response_body = UpdateIndexResponse {
+        name: body.name,
+        uid: index_uid,
+        created_at,
+        updated_at,
+    };
+
+    Ok(tide::response::json(response_body)
+        .with_status(StatusCode::ACCEPTED)
+        .into_response())
+}
+
+#[derive(Default, Deserialize)]
+#[serde(rename_all = "camelCase", deny_unknown_fields)]
+struct SchemaParams {
+    raw: bool,
+}
+
+pub async fn get_index_schema(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(IndexesRead)?;
+
+    let index = ctx.index()?;
+
+    // Tide doesn't support "no query param"
+    let params: SchemaParams = ctx.url_query().unwrap_or_default();
+
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+
+    let schema = index
+        .main
+        .schema(&reader)
+        .map_err(ResponseError::open_index)?;
+
+    match schema {
+        Some(schema) => {
+            if params.raw {
+                Ok(tide::response::json(schema))
+            } else {
+                Ok(tide::response::json(SchemaBody::from(schema)))
+            }
+        }
+        None => Err(ResponseError::not_found("missing index schema")),
+    }
+}
+
+pub async fn update_schema(mut ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(IndexesWrite)?;
+
+    let index_uid = ctx.url_param("index")?;
+
+    let params: SchemaParams = ctx.url_query().unwrap_or_default();
+
+    let schema = if params.raw {
+        ctx.body_json::<SchemaBuilder>()
+            .await
+            .map_err(ResponseError::bad_request)?
+            .build()
+    } else {
+        ctx.body_json::<SchemaBody>()
+            .await
+            .map_err(ResponseError::bad_request)?
+            .into()
+    };
+
+    let db = &ctx.state().db;
+    let mut writer = db.update_write_txn().map_err(ResponseError::internal)?;
+
+    let index = db
+        .open_index(&index_uid)
+        .ok_or(ResponseError::index_not_found(index_uid))?;
+
+    let update_id = index
+        .schema_update(&mut writer, schema.clone())
+        .map_err(ResponseError::internal)?;
+
+    writer.commit().map_err(ResponseError::internal)?;
+
+    let response_body = IndexUpdateResponse { update_id };
+    Ok(tide::response::json(response_body)
+        .with_status(StatusCode::ACCEPTED)
+        .into_response())
+}
+
+pub async fn get_update_status(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(IndexesRead)?;
+
+    let db = &ctx.state().db;
+    let reader = db.update_read_txn().map_err(ResponseError::internal)?;
+
+    let update_id = ctx
+        .param::<u64>("update_id")
+        .map_err(|e| ResponseError::bad_parameter("update_id", e))?;
+
+    let index = ctx.index()?;
+    let status = index
+        .update_status(&reader, update_id)
+        .map_err(ResponseError::internal)?;
+
+    let response = match status {
+        Some(status) => tide::response::json(status)
+            .with_status(StatusCode::OK)
+            .into_response(),
+        None => tide::response::json(json!({ "message": "unknown update id" }))
+            .with_status(StatusCode::NOT_FOUND)
+            .into_response(),
+    };
+
+    Ok(response)
+}
+
+pub async fn get_all_updates_status(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(IndexesRead)?;
+
+    let db = &ctx.state().db;
+    let reader = db.update_read_txn().map_err(ResponseError::internal)?;
+
+    let index = ctx.index()?;
+    let all_status = index
+        .all_updates_status(&reader)
+        .map_err(ResponseError::internal)?;
+
+    let response = tide::response::json(all_status)
+        .with_status(StatusCode::OK)
+        .into_response();
+
+    Ok(response)
+}
+
+pub async fn delete_index(ctx: Context<Data>) -> SResult<StatusCode> {
+    ctx.is_allowed(IndexesWrite)?;
+    let index_uid = ctx.url_param("index")?;
+
+    let found = ctx
+        .state()
+        .db
+        .delete_index(&index_uid)
+        .map_err(ResponseError::internal)?;
+
+    if found {
+        Ok(StatusCode::NO_CONTENT)
+    } else {
+        Ok(StatusCode::NOT_FOUND)
+    }
+}
+
+pub fn index_update_callback(index_uid: &str, data: &Data, status: ProcessedUpdateResult) {
+    if status.error.is_some() {
+        return;
+    }
+
+    if let Some(index) = data.db.open_index(&index_uid) {
+        let db = &data.db;
+        let mut writer = match db.main_write_txn() {
+            Ok(writer) => writer,
+            Err(e) => {
+                error!("Impossible to get write_txn; {}", e);
+                return;
+            }
+        };
+
+        if let Err(e) = data.compute_stats(&mut writer, &index_uid) {
+            error!("Impossible to compute stats; {}", e)
+        }
+
+        if let Err(e) = data.set_last_update(&mut writer) {
+            error!("Impossible to update last_update; {}", e)
+        }
+
+        if let Err(e) = index.main.put_updated_at(&mut writer) {
+            error!("Impossible to update updated_at; {}", e)
+        }
+
+        if let Err(e) = writer.commit() {
+            error!("Impossible to get write_txn; {}", e);
+        }
+    }
+}

meilisearch-http/src/routes/key.rs

@@ -0,0 +1,194 @@
+use chrono::serde::ts_seconds;
+use chrono::{DateTime, Utc};
+use heed::types::{SerdeBincode, Str};
+use http::StatusCode;
+use rand::seq::SliceRandom;
+use serde::{Deserialize, Serialize};
+use tide::response::IntoResponse;
+use tide::{Context, Response};
+
+use crate::error::{ResponseError, SResult};
+use crate::helpers::tide::ContextExt;
+use crate::models::token::ACL::*;
+use crate::models::token::*;
+use crate::Data;
+
+fn generate_api_key() -> String {
+    let mut rng = rand::thread_rng();
+    let sample = b"ABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789";
+    sample
+        .choose_multiple(&mut rng, 40)
+        .map(|c| *c as char)
+        .collect()
+}
+
+pub async fn list(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(Admin)?;
+
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+
+    let common_store = db.common_store();
+
+    let mut response: Vec<Token> = Vec::new();
+
+    let iter = common_store
+        .prefix_iter::<_, Str, SerdeBincode<Token>>(&reader, TOKEN_PREFIX_KEY)
+        .map_err(ResponseError::internal)?;
+
+    for result in iter {
+        let (_, token) = result.map_err(ResponseError::internal)?;
+        response.push(token);
+    }
+
+    Ok(tide::response::json(response))
+}
+
+pub async fn get(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(Admin)?;
+    let request_key = ctx.url_param("key")?;
+
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+
+    let token_key = format!("{}{}", TOKEN_PREFIX_KEY, request_key);
+
+    let token_config = db
+        .common_store()
+        .get::<_, Str, SerdeBincode<Token>>(&reader, &token_key)
+        .map_err(ResponseError::internal)?
+        .ok_or(ResponseError::not_found(format!(
+            "token key: {}",
+            token_key
+        )))?;
+
+    Ok(tide::response::json(token_config))
+}
+
+#[derive(Debug, Serialize, Deserialize)]
+#[serde(rename_all = "camelCase", deny_unknown_fields)]
+pub struct CreatedRequest {
+    description: String,
+    acl: Vec<ACL>,
+    indexes: Vec<Wildcard>,
+    #[serde(with = "ts_seconds")]
+    expires_at: DateTime<Utc>,
+}
+
+pub async fn create(mut ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(Admin)?;
+
+    let data: CreatedRequest = ctx.body_json().await.map_err(ResponseError::bad_request)?;
+
+    let key = generate_api_key();
+    let token_key = format!("{}{}", TOKEN_PREFIX_KEY, key);
+
+    let token_definition = Token {
+        key,
+        description: data.description,
+        acl: data.acl,
+        indexes: data.indexes,
+        expires_at: data.expires_at,
+        created_at: Utc::now(),
+        updated_at: Utc::now(),
+        revoked: false,
+    };
+
+    let db = &ctx.state().db;
+    let mut writer = db.main_write_txn().map_err(ResponseError::internal)?;
+
+    db.common_store()
+        .put::<_, Str, SerdeBincode<Token>>(&mut writer, &token_key, &token_definition)
+        .map_err(ResponseError::internal)?;
+
+    writer.commit().map_err(ResponseError::internal)?;
+
+    Ok(tide::response::json(token_definition)
+        .with_status(StatusCode::CREATED)
+        .into_response())
+}
+
+#[derive(Debug, Serialize, Deserialize)]
+#[serde(rename_all = "camelCase", deny_unknown_fields)]
+pub struct UpdatedRequest {
+    description: Option<String>,
+    acl: Option<Vec<ACL>>,
+    indexes: Option<Vec<Wildcard>>,
+    expires_at: Option<DateTime<Utc>>,
+    revoked: Option<bool>,
+}
+
+pub async fn update(mut ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(Admin)?;
+    let request_key = ctx.url_param("key")?;
+
+    let data: UpdatedRequest = ctx.body_json().await.map_err(ResponseError::bad_request)?;
+
+    let db = &ctx.state().db;
+    let mut writer = db.main_write_txn().map_err(ResponseError::internal)?;
+
+    let common_store = db.common_store();
+
+    let token_key = format!("{}{}", TOKEN_PREFIX_KEY, request_key);
+
+    let mut token_config = common_store
+        .get::<_, Str, SerdeBincode<Token>>(&writer, &token_key)
+        .map_err(ResponseError::internal)?
+        .ok_or(ResponseError::not_found(format!(
+            "token key: {}",
+            token_key
+        )))?;
+
+    // apply the modifications
+    if let Some(description) = data.description {
+        token_config.description = description;
+    }
+
+    if let Some(acl) = data.acl {
+        token_config.acl = acl;
+    }
+
+    if let Some(indexes) = data.indexes {
+        token_config.indexes = indexes;
+    }
+
+    if let Some(expires_at) = data.expires_at {
+        token_config.expires_at = expires_at;
+    }
+
+    if let Some(revoked) = data.revoked {
+        token_config.revoked = revoked;
+    }
+
+    token_config.updated_at = Utc::now();
+
+    common_store
+        .put::<_, Str, SerdeBincode<Token>>(&mut writer, &token_key, &token_config)
+        .map_err(ResponseError::internal)?;
+
+    writer.commit().map_err(ResponseError::internal)?;
+
+    Ok(tide::response::json(token_config)
+        .with_status(StatusCode::OK)
+        .into_response())
+}
+
+pub async fn delete(ctx: Context<Data>) -> SResult<StatusCode> {
+    ctx.is_allowed(Admin)?;
+    let request_key = ctx.url_param("key")?;
+
+    let db = &ctx.state().db;
+    let mut writer = db.main_write_txn().map_err(ResponseError::internal)?;
+
+    let common_store = db.common_store();
+
+    let token_key = format!("{}{}", TOKEN_PREFIX_KEY, request_key);
+
+    common_store
+        .delete::<_, Str>(&mut writer, &token_key)
+        .map_err(ResponseError::internal)?;
+
+    writer.commit().map_err(ResponseError::internal)?;
+
+    Ok(StatusCode::NO_CONTENT)
+}

meilisearch-http/src/routes/mod.rs

@@ -0,0 +1,122 @@
+use crate::data::Data;
+
+pub mod document;
+pub mod health;
+pub mod index;
+pub mod key;
+pub mod search;
+pub mod setting;
+pub mod stats;
+pub mod stop_words;
+pub mod synonym;
+
+pub fn load_routes(app: &mut tide::App<Data>) {
+    app.at("").nest(|router| {
+        router.at("/indexes").nest(|router| {
+            router
+                .at("/")
+                .get(index::list_indexes)
+                .post(index::create_index);
+
+            router.at("/search").post(search::search_multi_index);
+
+            router.at("/:index").nest(|router| {
+                router.at("/search").get(search::search_with_url_query);
+
+                router.at("/updates").nest(|router| {
+                    router.at("/").get(index::get_all_updates_status);
+
+                    router.at("/:update_id").get(index::get_update_status);
+                });
+
+                router
+                    .at("/")
+                    .get(index::get_index)
+                    .put(index::update_index)
+                    .delete(index::delete_index);
+
+                router
+                    .at("/schema")
+                    .get(index::get_index_schema)
+                    .put(index::update_schema);
+
+                router.at("/documents").nest(|router| {
+                    router
+                        .at("/")
+                        .get(document::get_all_documents)
+                        .post(document::add_or_replace_multiple_documents)
+                        .put(document::add_or_update_multiple_documents)
+                        .delete(document::clear_all_documents);
+
+                    router.at("/:identifier").nest(|router| {
+                        router
+                            .at("/")
+                            .get(document::get_document)
+                            .delete(document::delete_document);
+                    });
+
+                    router
+                        .at("/delete")
+                        .post(document::delete_multiple_documents);
+                });
+
+                router.at("/synonyms").nest(|router| {
+                    router
+                        .at("/")
+                        .get(synonym::list)
+                        .post(synonym::create)
+                        .delete(synonym::clear);
+
+                    router
+                        .at("/:synonym")
+                        .get(synonym::get)
+                        .put(synonym::update)
+                        .delete(synonym::delete);
+
+                    router.at("/batch").post(synonym::batch_write);
+                });
+
+                router.at("/stop-words").nest(|router| {
+                    router
+                        .at("/")
+                        .get(stop_words::list)
+                        .patch(stop_words::add)
+                        .delete(stop_words::delete);
+                });
+
+                router
+                    .at("/settings")
+                    .get(setting::get)
+                    .post(setting::update);
+            });
+        });
+
+        router.at("/keys").nest(|router| {
+            router.at("/").get(key::list).post(key::create);
+
+            router
+                .at("/:key")
+                .get(key::get)
+                .put(key::update)
+                .delete(key::delete);
+        });
+    });
+
+    // Private
+    app.at("").nest(|router| {
+        router
+            .at("/health")
+            .get(health::get_health)
+            .post(health::set_healthy)
+            .put(health::change_healthyness)
+            .delete(health::set_unhealthy);
+
+        router.at("/stats").get(stats::get_stats);
+        router.at("/stats/:index").get(stats::index_stat);
+        router.at("/version").get(stats::get_version);
+        router.at("/sys-info").get(stats::get_sys_info);
+        router
+            .at("/sys-info/pretty")
+            .get(stats::get_sys_info_pretty);
+    });
+}

meilisearch-http/src/routes/search.rs

@@ -0,0 +1,243 @@
+use std::collections::HashMap;
+use std::collections::HashSet;
+use std::time::Duration;
+
+use meilisearch_core::Index;
+use rayon::iter::{IntoParallelIterator, ParallelIterator};
+use serde::{Deserialize, Serialize};
+use tide::querystring::ContextExt as QSContextExt;
+use tide::{Context, Response};
+
+use crate::error::{ResponseError, SResult};
+use crate::helpers::meilisearch::{Error, IndexSearchExt, SearchHit};
+use crate::helpers::tide::ContextExt;
+use crate::Data;
+
+#[derive(Deserialize)]
+#[serde(rename_all = "camelCase", deny_unknown_fields)]
+struct SearchQuery {
+    q: String,
+    offset: Option<usize>,
+    limit: Option<usize>,
+    attributes_to_retrieve: Option<String>,
+    attributes_to_search_in: Option<String>,
+    attributes_to_crop: Option<String>,
+    crop_length: Option<usize>,
+    attributes_to_highlight: Option<String>,
+    filters: Option<String>,
+    timeout_ms: Option<u64>,
+    matches: Option<bool>,
+}
+
+pub async fn search_with_url_query(ctx: Context<Data>) -> SResult<Response> {
+    // ctx.is_allowed(DocumentsRead)?;
+
+    let index = ctx.index()?;
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+
+    let schema = index
+        .main
+        .schema(&reader)
+        .map_err(ResponseError::internal)?
+        .ok_or(ResponseError::open_index("No Schema found"))?;
+
+    let query: SearchQuery = ctx
+        .url_query()
+        .map_err(|_| ResponseError::bad_request("invalid query parameter"))?;
+
+    let mut search_builder = index.new_search(query.q.clone());
+
+    if let Some(offset) = query.offset {
+        search_builder.offset(offset);
+    }
+    if let Some(limit) = query.limit {
+        search_builder.limit(limit);
+    }
+
+    if let Some(attributes_to_retrieve) = query.attributes_to_retrieve {
+        for attr in attributes_to_retrieve.split(',') {
+            search_builder.add_retrievable_field(attr.to_string());
+        }
+    }
+    if let Some(attributes_to_search_in) = query.attributes_to_search_in {
+        for attr in attributes_to_search_in.split(',') {
+            search_builder.add_attribute_to_search_in(attr.to_string());
+        }
+    }
+    if let Some(attributes_to_crop) = query.attributes_to_crop {
+        let crop_length = query.crop_length.unwrap_or(200);
+        if attributes_to_crop == "*" {
+            let attributes_to_crop = schema
+                .iter()
+                .map(|(attr, ..)| (attr.to_string(), crop_length))
+                .collect();
+            search_builder.attributes_to_crop(attributes_to_crop);
+        } else {
+            let attributes_to_crop = attributes_to_crop
+                .split(',')
+                .map(|r| (r.to_string(), crop_length))
+                .collect();
+            search_builder.attributes_to_crop(attributes_to_crop);
+        }
+    }
+
+    if let Some(attributes_to_highlight) = query.attributes_to_highlight {
+        let attributes_to_highlight = if attributes_to_highlight == "*" {
+            schema.iter().map(|(attr, ..)| attr.to_string()).collect()
+        } else {
+            attributes_to_highlight
+                .split(',')
+                .map(ToString::to_string)
+                .collect()
+        };
+
+        search_builder.attributes_to_highlight(attributes_to_highlight);
+    }
+
+    if let Some(filters) = query.filters {
+        search_builder.filters(filters);
+    }
+
+    if let Some(timeout_ms) = query.timeout_ms {
+        search_builder.timeout(Duration::from_millis(timeout_ms));
+    }
+
+    if let Some(matches) = query.matches {
+        if matches {
+            search_builder.get_matches();
+        }
+    }
+
+    let response = match search_builder.search(&reader) {
+        Ok(response) => response,
+        Err(Error::Internal(message)) => return Err(ResponseError::Internal(message)),
+        Err(others) => return Err(ResponseError::bad_request(others)),
+    };
+
+    Ok(tide::response::json(response))
+}
+
+#[derive(Clone, Deserialize)]
+#[serde(rename_all = "camelCase", deny_unknown_fields)]
+struct SearchMultiBody {
+    indexes: HashSet<String>,
+    query: String,
+    offset: Option<usize>,
+    limit: Option<usize>,
+    attributes_to_retrieve: Option<HashSet<String>>,
+    attributes_to_search_in: Option<HashSet<String>>,
+    attributes_to_crop: Option<HashMap<String, usize>>,
+    attributes_to_highlight: Option<HashSet<String>>,
+    filters: Option<String>,
+    timeout_ms: Option<u64>,
+    matches: Option<bool>,
+}
+
+#[derive(Debug, Clone, Serialize)]
+#[serde(rename_all = "camelCase")]
+struct SearchMultiBodyResponse {
+    hits: HashMap<String, Vec<SearchHit>>,
+    offset: usize,
+    hits_per_page: usize,
+    processing_time_ms: usize,
+    query: String,
+}
+
+pub async fn search_multi_index(mut ctx: Context<Data>) -> SResult<Response> {
+    // ctx.is_allowed(DocumentsRead)?;
+    let body = ctx
+        .body_json::<SearchMultiBody>()
+        .await
+        .map_err(ResponseError::bad_request)?;
+
+    let mut index_list = body.clone().indexes;
+
+    for index in index_list.clone() {
+        if index == "*" {
+            index_list = ctx.state().db.indexes_uids().into_iter().collect();
+            break;
+        }
+    }
+
+    let mut offset = 0;
+    let mut count = 20;
+
+    if let Some(body_offset) = body.offset {
+        if let Some(limit) = body.limit {
+            offset = body_offset;
+            count = limit;
+        }
+    }
+
+    let offset = offset;
+    let count = count;
+    let db = &ctx.state().db;
+    let par_body = body.clone();
+    let responses_per_index: Vec<SResult<_>> = index_list
+        .into_par_iter()
+        .map(move |index_uid| {
+            let index: Index = db
+                .open_index(&index_uid)
+                .ok_or(ResponseError::index_not_found(&index_uid))?;
+
+            let mut search_builder = index.new_search(par_body.query.clone());
+
+            search_builder.offset(offset);
+            search_builder.limit(count);
+
+            if let Some(attributes_to_retrieve) = par_body.attributes_to_retrieve.clone() {
+                search_builder.attributes_to_retrieve(attributes_to_retrieve);
+            }
+            if let Some(attributes_to_search_in) = par_body.attributes_to_search_in.clone() {
+                search_builder.attributes_to_search_in(attributes_to_search_in);
+            }
+            if let Some(attributes_to_crop) = par_body.attributes_to_crop.clone() {
+                search_builder.attributes_to_crop(attributes_to_crop);
+            }
+            if let Some(attributes_to_highlight) = par_body.attributes_to_highlight.clone() {
+                search_builder.attributes_to_highlight(attributes_to_highlight);
+            }
+            if let Some(filters) = par_body.filters.clone() {
+                search_builder.filters(filters);
+            }
+            if let Some(timeout_ms) = par_body.timeout_ms {
+                search_builder.timeout(Duration::from_millis(timeout_ms));
+            }
+            if let Some(matches) = par_body.matches {
+                if matches {
+                    search_builder.get_matches();
+                }
+            }
+
+            let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+            let response = search_builder
+                .search(&reader)
+                .map_err(ResponseError::internal)?;
+            Ok((index_uid, response))
+        })
+        .collect();
+
+    let mut hits_map = HashMap::new();
+
+    let mut max_query_time = 0;
+
+    for response in responses_per_index {
+        if let Ok((index_uid, response)) = response {
+            if response.processing_time_ms > max_query_time {
+                max_query_time = response.processing_time_ms;
+            }
+            hits_map.insert(index_uid, response.hits);
+        }
+    }
+
+    let response = SearchMultiBodyResponse {
+        hits: hits_map,
+        offset,
+        hits_per_page: count,
+        processing_time_ms: max_query_time,
+        query: body.query,
+    };
+
+    Ok(tide::response::json(response))
+}

meilisearch-http/src/routes/setting.rs

@@ -0,0 +1,88 @@
+use std::collections::HashMap;
+
+use http::StatusCode;
+use serde::{Deserialize, Serialize};
+use tide::response::IntoResponse;
+use tide::{Context, Response};
+
+use crate::error::{ResponseError, SResult};
+use crate::helpers::tide::ContextExt;
+use crate::models::token::ACL::*;
+use crate::routes::document::IndexUpdateResponse;
+use crate::Data;
+
+#[derive(Default, Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
+#[serde(rename_all = "camelCase", deny_unknown_fields)]
+pub struct SettingBody {
+    pub ranking_order: Option<RankingOrder>,
+    pub distinct_field: Option<DistinctField>,
+    pub ranking_rules: Option<RankingRules>,
+}
+
+#[derive(Debug, Copy, Clone, PartialEq, Eq, Serialize, Deserialize)]
+#[serde(rename_all = "lowercase")]
+pub enum RankingOrdering {
+    Asc,
+    Dsc,
+}
+
+pub type RankingOrder = Vec<String>;
+pub type DistinctField = String;
+pub type RankingRules = HashMap<String, RankingOrdering>;
+
+pub async fn get(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(SettingsRead)?;
+    let index = ctx.index()?;
+
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+
+    let settings = match index.main.customs(&reader).unwrap() {
+        Some(bytes) => bincode::deserialize(bytes).unwrap(),
+        None => SettingBody::default(),
+    };
+
+    Ok(tide::response::json(settings))
+}
+
+pub async fn update(mut ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(SettingsWrite)?;
+
+    let settings: SettingBody = ctx.body_json().await.map_err(ResponseError::bad_request)?;
+
+    let index = ctx.index()?;
+
+    let db = &ctx.state().db;
+    let reader = db.main_write_txn().map_err(ResponseError::internal)?;
+    let mut writer = db.update_write_txn().map_err(ResponseError::internal)?;
+
+    let mut current_settings = match index.main.customs(&reader).unwrap() {
+        Some(bytes) => bincode::deserialize(bytes).unwrap(),
+        None => SettingBody::default(),
+    };
+
+    if let Some(ranking_order) = settings.ranking_order {
+        current_settings.ranking_order = Some(ranking_order);
+    }
+
+    if let Some(distinct_field) = settings.distinct_field {
+        current_settings.distinct_field = Some(distinct_field);
+    }
+
+    if let Some(ranking_rules) = settings.ranking_rules {
+        current_settings.ranking_rules = Some(ranking_rules);
+    }
+
+    let bytes = bincode::serialize(&current_settings).unwrap();
+
+    let update_id = index
+        .customs_update(&mut writer, bytes)
+        .map_err(ResponseError::internal)?;
+
+    writer.commit().map_err(ResponseError::internal)?;
+
+    let response_body = IndexUpdateResponse { update_id };
+    Ok(tide::response::json(response_body)
+        .with_status(StatusCode::ACCEPTED)
+        .into_response())
+}

meilisearch-http/src/routes/stats.rs

@@ -0,0 +1,338 @@
+use std::collections::HashMap;
+
+use chrono::{DateTime, Utc};
+use log::error;
+use pretty_bytes::converter::convert;
+use serde::Serialize;
+use sysinfo::{NetworkExt, Pid, ProcessExt, ProcessorExt, System, SystemExt};
+use tide::{Context, Response};
+use walkdir::WalkDir;
+
+use crate::error::{ResponseError, SResult};
+use crate::helpers::tide::ContextExt;
+use crate::models::token::ACL::*;
+use crate::Data;
+
+#[derive(Serialize)]
+#[serde(rename_all = "camelCase")]
+struct IndexStatsResponse {
+    number_of_documents: u64,
+    is_indexing: bool,
+    fields_frequency: HashMap<String, usize>,
+}
+
+pub async fn index_stat(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(Admin)?;
+    let index_uid = ctx.url_param("index")?;
+    let index = ctx.index()?;
+
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+    let update_reader = db.update_read_txn().map_err(ResponseError::internal)?;
+
+    let number_of_documents = index
+        .main
+        .number_of_documents(&reader)
+        .map_err(ResponseError::internal)?;
+
+    let fields_frequency = index
+        .main
+        .fields_frequency(&reader)
+        .map_err(ResponseError::internal)?
+        .unwrap_or_default();
+
+    let is_indexing = ctx
+        .state()
+        .is_indexing(&update_reader, &index_uid)
+        .map_err(ResponseError::internal)?
+        .ok_or(ResponseError::internal("'is_indexing' date not found"))?;
+
+    let response = IndexStatsResponse {
+        number_of_documents,
+        is_indexing,
+        fields_frequency,
+    };
+    Ok(tide::response::json(response))
+}
+
+#[derive(Serialize)]
+#[serde(rename_all = "camelCase")]
+struct StatsResult {
+    database_size: u64,
+    last_update: Option<DateTime<Utc>>,
+    indexes: HashMap<String, IndexStatsResponse>,
+}
+
+pub async fn get_stats(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(Admin)?;
+
+    let mut index_list = HashMap::new();
+
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+    let update_reader = db.update_read_txn().map_err(ResponseError::internal)?;
+
+    let indexes_set = ctx.state().db.indexes_uids();
+    for index_uid in indexes_set {
+        let index = ctx.state().db.open_index(&index_uid);
+
+        match index {
+            Some(index) => {
+                let number_of_documents = index
+                    .main
+                    .number_of_documents(&reader)
+                    .map_err(ResponseError::internal)?;
+
+                let fields_frequency = index
+                    .main
+                    .fields_frequency(&reader)
+                    .map_err(ResponseError::internal)?
+                    .unwrap_or_default();
+
+                let is_indexing = ctx
+                    .state()
+                    .is_indexing(&update_reader, &index_uid)
+                    .map_err(ResponseError::internal)?
+                    .ok_or(ResponseError::internal("'is_indexing' date not found"))?;
+
+                let response = IndexStatsResponse {
+                    number_of_documents,
+                    is_indexing,
+                    fields_frequency,
+                };
+                index_list.insert(index_uid, response);
+            }
+            None => error!(
+                "Index {:?} is referenced in the indexes list but cannot be found",
+                index_uid
+            ),
+        }
+    }
+
+    let database_size = WalkDir::new(ctx.state().db_path.clone())
+        .into_iter()
+        .filter_map(|entry| entry.ok())
+        .filter_map(|entry| entry.metadata().ok())
+        .filter(|metadata| metadata.is_file())
+        .fold(0, |acc, m| acc + m.len());
+
+    let last_update = ctx
+        .state()
+        .last_update(&reader)
+        .map_err(ResponseError::internal)?;
+
+    let response = StatsResult {
+        database_size,
+        last_update,
+        indexes: index_list,
+    };
+
+    Ok(tide::response::json(response))
+}
+
+#[derive(Serialize)]
+#[serde(rename_all = "camelCase")]
+struct VersionResponse {
+    commit_sha: String,
+    build_date: String,
+    pkg_version: String,
+}
+
+pub async fn get_version(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(Admin)?;
+    let response = VersionResponse {
+        commit_sha: env!("VERGEN_SHA").to_string(),
+        build_date: env!("VERGEN_BUILD_TIMESTAMP").to_string(),
+        pkg_version: env!("CARGO_PKG_VERSION").to_string(),
+    };
+
+    Ok(tide::response::json(response))
+}
+
+#[derive(Serialize)]
+#[serde(rename_all = "camelCase")]
+pub(crate) struct SysGlobal {
+    total_memory: u64,
+    used_memory: u64,
+    total_swap: u64,
+    used_swap: u64,
+    input_data: u64,
+    output_data: u64,
+}
+
+impl SysGlobal {
+    fn new() -> SysGlobal {
+        SysGlobal {
+            total_memory: 0,
+            used_memory: 0,
+            total_swap: 0,
+            used_swap: 0,
+            input_data: 0,
+            output_data: 0,
+        }
+    }
+}
+
+#[derive(Serialize)]
+#[serde(rename_all = "camelCase")]
+pub(crate) struct SysProcess {
+    memory: u64,
+    cpu: f32,
+}
+
+impl SysProcess {
+    fn new() -> SysProcess {
+        SysProcess {
+            memory: 0,
+            cpu: 0.0,
+        }
+    }
+}
+
+#[derive(Serialize)]
+#[serde(rename_all = "camelCase")]
+pub(crate) struct SysInfo {
+    memory_usage: f64,
+    processor_usage: Vec<f32>,
+    global: SysGlobal,
+    process: SysProcess,
+}
+
+impl SysInfo {
+    fn new() -> SysInfo {
+        SysInfo {
+            memory_usage: 0.0,
+            processor_usage: Vec::new(),
+            global: SysGlobal::new(),
+            process: SysProcess::new(),
+        }
+    }
+}
+
+pub(crate) fn report(pid: Pid) -> SysInfo {
+    let mut sys = System::new();
+    let mut info = SysInfo::new();
+
+    info.memory_usage = sys.get_used_memory() as f64 / sys.get_total_memory() as f64 * 100.0;
+
+    for processor in sys.get_processor_list() {
+        info.processor_usage.push(processor.get_cpu_usage() * 100.0);
+    }
+
+    info.global.total_memory = sys.get_total_memory();
+    info.global.used_memory = sys.get_used_memory();
+    info.global.total_swap = sys.get_total_swap();
+    info.global.used_swap = sys.get_used_swap();
+    info.global.input_data = sys.get_network().get_income();
+    info.global.output_data = sys.get_network().get_outcome();
+
+    if let Some(process) = sys.get_process(pid) {
+        info.process.memory = process.memory();
+        info.process.cpu = process.cpu_usage() * 100.0;
+    }
+
+    sys.refresh_all();
+
+    info
+}
+
+pub async fn get_sys_info(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(Admin)?;
+    Ok(tide::response::json(report(ctx.state().server_pid)))
+}
+
+#[derive(Serialize)]
+#[serde(rename_all = "camelCase")]
+pub(crate) struct SysGlobalPretty {
+    total_memory: String,
+    used_memory: String,
+    total_swap: String,
+    used_swap: String,
+    input_data: String,
+    output_data: String,
+}
+
+impl SysGlobalPretty {
+    fn new() -> SysGlobalPretty {
+        SysGlobalPretty {
+            total_memory: "None".to_owned(),
+            used_memory: "None".to_owned(),
+            total_swap: "None".to_owned(),
+            used_swap: "None".to_owned(),
+            input_data: "None".to_owned(),
+            output_data: "None".to_owned(),
+        }
+    }
+}
+
+#[derive(Serialize)]
+#[serde(rename_all = "camelCase")]
+pub(crate) struct SysProcessPretty {
+    memory: String,
+    cpu: String,
+}
+
+impl SysProcessPretty {
+    fn new() -> SysProcessPretty {
+        SysProcessPretty {
+            memory: "None".to_owned(),
+            cpu: "None".to_owned(),
+        }
+    }
+}
+
+#[derive(Serialize)]
+#[serde(rename_all = "camelCase")]
+pub(crate) struct SysInfoPretty {
+    memory_usage: String,
+    processor_usage: Vec<String>,
+    global: SysGlobalPretty,
+    process: SysProcessPretty,
+}
+
+impl SysInfoPretty {
+    fn new() -> SysInfoPretty {
+        SysInfoPretty {
+            memory_usage: "None".to_owned(),
+            processor_usage: Vec::new(),
+            global: SysGlobalPretty::new(),
+            process: SysProcessPretty::new(),
+        }
+    }
+}
+
+pub(crate) fn report_pretty(pid: Pid) -> SysInfoPretty {
+    let mut sys = System::new();
+    let mut info = SysInfoPretty::new();
+
+    info.memory_usage = format!(
+        "{:.1} %",
+        sys.get_used_memory() as f64 / sys.get_total_memory() as f64 * 100.0
+    );
+
+    for processor in sys.get_processor_list() {
+        info.processor_usage
+            .push(format!("{:.1} %", processor.get_cpu_usage() * 100.0));
+    }
+
+    info.global.total_memory = convert(sys.get_total_memory() as f64 * 1024.0);
+    info.global.used_memory = convert(sys.get_used_memory() as f64 * 1024.0);
+    info.global.total_swap = convert(sys.get_total_swap() as f64 * 1024.0);
+    info.global.used_swap = convert(sys.get_used_swap() as f64 * 1024.0);
+    info.global.input_data = convert(sys.get_network().get_income() as f64);
+    info.global.output_data = convert(sys.get_network().get_outcome() as f64);
+
+    if let Some(process) = sys.get_process(pid) {
+        info.process.memory = convert(process.memory() as f64 * 1024.0);
+        info.process.cpu = format!("{:.1} %", process.cpu_usage() * 100.0);
+    }
+
+    sys.refresh_all();
+
+    info
+}
+
+pub async fn get_sys_info_pretty(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(Admin)?;
+    Ok(tide::response::json(report_pretty(ctx.state().server_pid)))
+}

meilisearch-http/src/routes/stop_words.rs

@@ -0,0 +1,82 @@
+use http::StatusCode;
+use tide::response::IntoResponse;
+use tide::{Context, Response};
+
+use crate::error::{ResponseError, SResult};
+use crate::helpers::tide::ContextExt;
+use crate::models::token::ACL::*;
+use crate::routes::document::IndexUpdateResponse;
+use crate::Data;
+
+pub async fn list(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(SettingsRead)?;
+    let index = ctx.index()?;
+
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+
+    let stop_words_fst = index
+        .main
+        .stop_words_fst(&reader)
+        .map_err(ResponseError::internal)?;
+
+    let stop_words = stop_words_fst
+        .unwrap_or_default()
+        .stream()
+        .into_strs()
+        .map_err(ResponseError::internal)?;
+
+    Ok(tide::response::json(stop_words))
+}
+
+pub async fn add(mut ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(SettingsRead)?;
+    let index = ctx.index()?;
+
+    let data: Vec<String> = ctx.body_json().await.map_err(ResponseError::bad_request)?;
+
+    let db = &ctx.state().db;
+    let mut writer = db.update_write_txn().map_err(ResponseError::internal)?;
+
+    let mut stop_words_addition = index.stop_words_addition();
+    for stop_word in data {
+        stop_words_addition.add_stop_word(stop_word);
+    }
+
+    let update_id = stop_words_addition
+        .finalize(&mut writer)
+        .map_err(ResponseError::internal)?;
+
+    writer.commit().map_err(ResponseError::internal)?;
+
+    let response_body = IndexUpdateResponse { update_id };
+    Ok(tide::response::json(response_body)
+        .with_status(StatusCode::ACCEPTED)
+        .into_response())
+}
+
+pub async fn delete(mut ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(SettingsRead)?;
+    let index = ctx.index()?;
+
+    let data: Vec<String> = ctx.body_json().await.map_err(ResponseError::bad_request)?;
+
+    let db = &ctx.state().db;
+    let mut writer = db.update_write_txn().map_err(ResponseError::internal)?;
+
+    let mut stop_words_deletion = index.stop_words_deletion();
+    for stop_word in data {
+        stop_words_deletion.delete_stop_word(stop_word);
+    }
+
+    let update_id = stop_words_deletion
+        .finalize(&mut writer)
+        .map_err(ResponseError::internal)?;
+
+    writer.commit().map_err(ResponseError::internal)?;
+
+    let response_body = IndexUpdateResponse { update_id };
+    Ok(tide::response::json(response_body)
+        .with_status(StatusCode::ACCEPTED)
+        .into_response())
+}

meilisearch-http/src/routes/synonym.rs

@@ -0,0 +1,236 @@
+use std::collections::HashMap;
+
+use http::StatusCode;
+use serde::{Deserialize, Serialize};
+use tide::response::IntoResponse;
+use tide::{Context, Response};
+
+use crate::error::{ResponseError, SResult};
+use crate::helpers::tide::ContextExt;
+use crate::models::token::ACL::*;
+use crate::routes::document::IndexUpdateResponse;
+use crate::Data;
+
+#[derive(Clone, Serialize, Deserialize)]
+#[serde(untagged)]
+pub enum Synonym {
+    OneWay(SynonymOneWay),
+    MultiWay { synonyms: Vec<String> },
+}
+
+#[derive(Clone, Serialize, Deserialize)]
+#[serde(rename_all = "camelCase", deny_unknown_fields)]
+pub struct SynonymOneWay {
+    pub input: String,
+    pub synonyms: Vec<String>,
+}
+
+pub type Synonyms = Vec<Synonym>;
+
+pub async fn list(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(SettingsRead)?;
+    let index = ctx.index()?;
+
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+
+    let synonyms_fst = index
+        .main
+        .synonyms_fst(&reader)
+        .map_err(ResponseError::internal)?;
+
+    let synonyms_fst = synonyms_fst.unwrap_or_default();
+    let synonyms_list = synonyms_fst.stream().into_strs().unwrap();
+
+    let mut response = HashMap::new();
+
+    let index_synonyms = &index.synonyms;
+
+    for synonym in synonyms_list {
+        let alternative_list = index_synonyms
+            .synonyms(&reader, synonym.as_bytes())
+            .unwrap()
+            .unwrap()
+            .stream()
+            .into_strs()
+            .unwrap();
+        response.insert(synonym, alternative_list);
+    }
+
+    Ok(tide::response::json(response))
+}
+
+pub async fn get(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(SettingsRead)?;
+    let synonym = ctx.url_param("synonym")?;
+    let index = ctx.index()?;
+
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+
+    let synonym_list = index
+        .synonyms
+        .synonyms(&reader, synonym.as_bytes())
+        .unwrap()
+        .unwrap()
+        .stream()
+        .into_strs()
+        .unwrap();
+
+    Ok(tide::response::json(synonym_list))
+}
+
+pub async fn create(mut ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(SettingsWrite)?;
+
+    let data: Synonym = ctx.body_json().await.map_err(ResponseError::bad_request)?;
+
+    let index = ctx.index()?;
+
+    let db = &ctx.state().db;
+    let mut writer = db.update_write_txn().map_err(ResponseError::internal)?;
+
+    let mut synonyms_addition = index.synonyms_addition();
+
+    match data.clone() {
+        Synonym::OneWay(content) => {
+            synonyms_addition.add_synonym(content.input, content.synonyms.into_iter())
+        }
+        Synonym::MultiWay { mut synonyms } => {
+            if synonyms.len() > 1 {
+                for _ in 0..synonyms.len() {
+                    let (first, elems) = synonyms.split_first().unwrap();
+                    synonyms_addition.add_synonym(first, elems.iter());
+                    synonyms.rotate_left(1);
+                }
+            }
+        }
+    }
+
+    let update_id = synonyms_addition
+        .finalize(&mut writer)
+        .map_err(ResponseError::internal)?;
+
+    writer.commit().map_err(ResponseError::internal)?;
+
+    let response_body = IndexUpdateResponse { update_id };
+    Ok(tide::response::json(response_body)
+        .with_status(StatusCode::ACCEPTED)
+        .into_response())
+}
+
+pub async fn update(mut ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(SettingsWrite)?;
+    let synonym = ctx.url_param("synonym")?;
+    let index = ctx.index()?;
+    let data: Vec<String> = ctx.body_json().await.map_err(ResponseError::bad_request)?;
+
+    let db = &ctx.state().db;
+    let mut writer = db.update_write_txn().map_err(ResponseError::internal)?;
+
+    let mut synonyms_addition = index.synonyms_addition();
+    synonyms_addition.add_synonym(synonym.clone(), data.clone().into_iter());
+    let update_id = synonyms_addition
+        .finalize(&mut writer)
+        .map_err(ResponseError::internal)?;
+
+    writer.commit().map_err(ResponseError::internal)?;
+
+    let response_body = IndexUpdateResponse { update_id };
+    Ok(tide::response::json(response_body)
+        .with_status(StatusCode::ACCEPTED)
+        .into_response())
+}
+
+pub async fn delete(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(SettingsWrite)?;
+    let synonym = ctx.url_param("synonym")?;
+    let index = ctx.index()?;
+
+    let db = &ctx.state().db;
+    let mut writer = db.update_write_txn().map_err(ResponseError::internal)?;
+
+    let mut synonyms_deletion = index.synonyms_deletion();
+    synonyms_deletion.delete_all_alternatives_of(synonym);
+    let update_id = synonyms_deletion
+        .finalize(&mut writer)
+        .map_err(ResponseError::internal)?;
+
+    writer.commit().map_err(ResponseError::internal)?;
+
+    let response_body = IndexUpdateResponse { update_id };
+    Ok(tide::response::json(response_body)
+        .with_status(StatusCode::ACCEPTED)
+        .into_response())
+}
+
+pub async fn batch_write(mut ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(SettingsWrite)?;
+
+    let data: Synonyms = ctx.body_json().await.map_err(ResponseError::bad_request)?;
+
+    let index = ctx.index()?;
+
+    let db = &ctx.state().db;
+    let mut writer = db.update_write_txn().map_err(ResponseError::internal)?;
+
+    let mut synonyms_addition = index.synonyms_addition();
+    for raw in data {
+        match raw {
+            Synonym::OneWay(content) => {
+                synonyms_addition.add_synonym(content.input, content.synonyms.into_iter())
+            }
+            Synonym::MultiWay { mut synonyms } => {
+                if synonyms.len() > 1 {
+                    for _ in 0..synonyms.len() {
+                        let (first, elems) = synonyms.split_first().unwrap();
+                        synonyms_addition.add_synonym(first, elems.iter());
+                        synonyms.rotate_left(1);
+                    }
+                }
+            }
+        }
+    }
+    let update_id = synonyms_addition
+        .finalize(&mut writer)
+        .map_err(ResponseError::internal)?;
+
+    writer.commit().map_err(ResponseError::internal)?;
+
+    let response_body = IndexUpdateResponse { update_id };
+    Ok(tide::response::json(response_body)
+        .with_status(StatusCode::ACCEPTED)
+        .into_response())
+}
+
+pub async fn clear(ctx: Context<Data>) -> SResult<Response> {
+    ctx.is_allowed(SettingsWrite)?;
+    let index = ctx.index()?;
+
+    let db = &ctx.state().db;
+    let reader = db.main_read_txn().map_err(ResponseError::internal)?;
+    let mut writer = db.update_write_txn().map_err(ResponseError::internal)?;
+
+    let synonyms_fst = index
+        .main
+        .synonyms_fst(&reader)
+        .map_err(ResponseError::internal)?;
+
+    let synonyms_fst = synonyms_fst.unwrap_or_default();
+    let synonyms_list = synonyms_fst.stream().into_strs().unwrap();
+
+    let mut synonyms_deletion = index.synonyms_deletion();
+    for synonym in synonyms_list {
+        synonyms_deletion.delete_all_alternatives_of(synonym);
+    }
+    let update_id = synonyms_deletion
+        .finalize(&mut writer)
+        .map_err(ResponseError::internal)?;
+
+    writer.commit().map_err(ResponseError::internal)?;
+
+    let response_body = IndexUpdateResponse { update_id };
+    Ok(tide::response::json(response_body)
+        .with_status(StatusCode::ACCEPTED)
+        .into_response())
+}

meilisearch-schema/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
-name = "meilidb-schema"
-version = "0.5.11"
+name = "meilisearch-schema"
+version = "0.8.1"
 authors = ["Kerollmops <renault.cle@gmail.com>"]
 edition = "2018"
 

meilisearch-schema/src/lib.rs

@@ -22,7 +22,7 @@ pub const RANKED: SchemaProps = SchemaProps {
     ranked: true,
 };
 
-#[derive(Debug, Copy, Clone, PartialEq, Eq, Serialize, Deserialize)]
+#[derive(Copy, Clone, PartialEq, Eq, Serialize, Deserialize)]
 pub struct SchemaProps {
     #[serde(default)]
     pub displayed: bool,
@@ -60,6 +60,36 @@ impl BitOr for SchemaProps {
     }
 }
 
+impl fmt::Debug for SchemaProps {
+    #[allow(non_camel_case_types)]
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        #[derive(Debug)]
+        struct DISPLAYED;
+
+        #[derive(Debug)]
+        struct INDEXED;
+
+        #[derive(Debug)]
+        struct RANKED;
+
+        let mut debug_set = f.debug_set();
+
+        if self.displayed {
+            debug_set.entry(&DISPLAYED);
+        }
+
+        if self.indexed {
+            debug_set.entry(&INDEXED);
+        }
+
+        if self.ranked {
+            debug_set.entry(&RANKED);
+        }
+
+        debug_set.finish()
+    }
+}
+
 #[derive(Serialize, Deserialize)]
 pub struct SchemaBuilder {
     identifier: String,
@@ -102,12 +132,12 @@ impl SchemaBuilder {
     }
 }
 
-#[derive(Debug, Clone, PartialEq, Eq)]
+#[derive(Clone, PartialEq, Eq)]
 pub struct Schema {
     inner: Arc<InnerSchema>,
 }
 
-#[derive(Debug, Clone, PartialEq, Eq)]
+#[derive(Clone, PartialEq, Eq)]
 struct InnerSchema {
     identifier: String,
     attrs: HashMap<String, SchemaAttr>,
@@ -139,6 +169,10 @@ impl Schema {
         attributes
     }
 
+    pub fn number_of_attributes(&self) -> usize {
+        self.inner.attrs.len()
+    }
+
     pub fn props(&self, attr: SchemaAttr) -> SchemaProps {
         let (_, props) = self.inner.props[attr.0 as usize];
         props
@@ -184,6 +218,16 @@ impl<'de> Deserialize<'de> for Schema {
     }
 }
 
+impl fmt::Debug for Schema {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        let builder = self.to_builder();
+        f.debug_struct("Schema")
+            .field("identifier", &builder.identifier)
+            .field("attributes", &builder.attributes)
+            .finish()
+    }
+}
+
 #[derive(Serialize, Deserialize, Debug, Copy, Clone, PartialOrd, Ord, PartialEq, Eq, Hash)]
 pub struct SchemaAttr(pub u16);
 
@@ -447,4 +491,43 @@ mod tests {
 
         Ok(())
     }
+
+    #[test]
+    fn debug_output() {
+        use std::fmt::Write as _;
+
+        let mut builder = SchemaBuilder::with_identifier("id");
+        builder.new_attribute("alpha", DISPLAYED);
+        builder.new_attribute("beta", DISPLAYED | INDEXED);
+        builder.new_attribute("gamma", INDEXED);
+        let schema = builder.build();
+
+        let mut output = String::new();
+        let _ = write!(&mut output, "{:#?}", schema);
+
+        let expected = r#"Schema {
+    identifier: "id",
+    attributes: {
+        "alpha": {
+            DISPLAYED,
+        },
+        "beta": {
+            DISPLAYED,
+            INDEXED,
+        },
+        "gamma": {
+            INDEXED,
+        },
+    },
+}"#;
+
+        assert_eq!(output, expected);
+
+        let mut output = String::new();
+        let _ = write!(&mut output, "{:?}", schema);
+
+        let expected = r#"Schema { identifier: "id", attributes: {"alpha": {DISPLAYED}, "beta": {DISPLAYED, INDEXED}, "gamma": {INDEXED}} }"#;
+
+        assert_eq!(output, expected);
+    }
 }

meilisearch-tokenizer/Cargo.toml

@@ -1,8 +1,9 @@
 [package]
-name = "meilidb-tokenizer"
-version = "0.5.11"
+name = "meilisearch-tokenizer"
+version = "0.8.1"
 authors = ["Kerollmops <renault.cle@gmail.com>"]
 edition = "2018"
 
 [dependencies]
+deunicode = "1.0.0"
 slice-group-by = "0.2.4"