4304: Add CUDA GPU support for Hugging Face embedders r=Kerollmops a=dureuill
Adds a "cuda" feature to `milli`.
Compiling with this feature requires that the CUDA support library be installed (see "with CUDA support" paragraph in https://huggingface.github.io/candle/guide/installation.html), and adds CUDA support to the `huggingFace` embedder.
To enable GPU support, users will need to:
1. Have a compatible NVidia GPU under Linux
2. Follow [the guide](https://huggingface.github.io/candle/guide/installation.html) to install the CUDA dependencies
3. Compile Meilisearch with the `cuda` feature: `cargo build --release --features cuda`
# Impact
Enabling the CUDA feature allows to use an available GPU to compute embeddings with a `huggingFace` embedder.
On an AWS Graviton 2, this yields a x3 - x5 improvement on indexing time.
# Technical details
- I had to change the CI so that the cuda feature is not included in the `Tests all features` workflow
- To achieve that, I had to add a binary following the `cargo xtask` design pattern, to list all features excepted the cuda one.
- I then changed the workflow accordingly (renamed to "Tests almost all features" 😉)
- A test run of the new feature was done on a temporary version of this PR that had it enabled for PRs: [See the results here](https://github.com/meilisearch/meilisearch/actions/runs/7461331929/job/20301216732)
Co-authored-by: Louis Dureuil <louis@meilisearch.com>
4238: Task queue webhook r=dureuill a=irevoire
# Prototype `prototype-task-queue-webhook-1`
The prototype is available through Docker by using the following command:
```bash
docker run -p 7700:7700 -v $(pwd)/meili_data:/meili_data getmeili/meilisearch:prototype-task-queue-webhook-1
```
# Pull Request
Implements the task queue webhook.
## Related issue
Fixes https://github.com/meilisearch/meilisearch/issues/4236
## What does this PR do?
- Provide a new cli and env var for the webhook, respectively called `--task-webhook-url` and `MEILI_TASK_WEBHOOK_URL`
- Also supports sending the requests with a custom `Authorization` header by specifying the optional `--task-webhook-authorization-header` CLI parameter or `MEILI_TASK_WEBHOOK_AUTHORIZATION_HEADER` env variable.
- Throw an error if the specified URL is invalid
- Every time a batch is processed, send all the finished tasks into the webhook with our public `TaskView` type as a JSON Line GZIPed body.
- Add one test.
## PR checklist
### Before becoming ready to review
- [x] Add a test
- [x] Compress the data we send
- [x] Chunk and stream the data we send
- [x] Remove the unwrap in the index-scheduler when sending the data fails
- [x] The analytics are missing
### Before merging
- [x] Release a prototype
Co-authored-by: Tamo <tamo@meilisearch.com>
Co-authored-by: Clément Renault <clement@meilisearch.com>
4009: Bump rustls-webpki from 0.100.1 to 0.100.2 r=Kerollmops a=dependabot[bot]
Bumps [rustls-webpki](https://github.com/rustls/webpki) from 0.100.1 to 0.100.2.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a href="https://github.com/rustls/webpki/releases">rustls-webpki's releases</a>.</em></p>
<blockquote>
<h2>v/0.100.2</h2>
<h2>Release notes</h2>
<ul>
<li>certificate path building and verification is now capped at 100 signature validation operations to avoid the risk of CPU usage denial-of-service attack when validating crafted certificate chains producing quadratic runtime. This risk affected both clients, as well as servers that verified client certificates.</li>
</ul>
<h2>What's Changed</h2>
<ul>
<li>v0.100.2 prep by <a href="https://github.com/cpu"><code>`@cpu</code></a>` in <a href="https://redirect.github.com/rustls/webpki/pull/154">rustls/webpki#154</a></li>
</ul>
<p><strong>Full Changelog</strong>: <a href="https://github.com/rustls/webpki/compare/v/0.100.1...v/0.100.2">https://github.com/rustls/webpki/compare/v/0.100.1...v/0.100.2</a></p>
</blockquote>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a href="c8b821450b"><code>c8b8214</code></a> Bump MSRV to 1.60</li>
<li><a href="855752292e"><code>8557522</code></a> Avoid testing MSRV of dev-dependencies</li>
<li><a href="73a7f0c7d7"><code>73a7f0c</code></a> Cargo: version 0.100.1 -> 0.100.2</li>
<li><a href="4ea052366f"><code>4ea0523</code></a> verify_cert: enforce maximum number of signatures.</li>
<li>See full diff in <a href="https://github.com/rustls/webpki/compare/v/0.100.1...v/0.100.2">compare view</a></li>
</ul>
</details>
<br />
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores)
Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting ``@dependabot` rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- ``@dependabot` rebase` will rebase this PR
- ``@dependabot` recreate` will recreate this PR, overwriting any edits that have been made to it
- ``@dependabot` merge` will merge this PR after your CI passes on it
- ``@dependabot` squash and merge` will squash and merge this PR after your CI passes on it
- ``@dependabot` cancel merge` will cancel a previously requested merge and block automerging
- ``@dependabot` reopen` will reopen this PR if it is closed
- ``@dependabot` close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
- ``@dependabot` show <dependency name> ignore conditions` will show all of the ignore conditions of the specified dependency
- ``@dependabot` ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
- ``@dependabot` ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
- ``@dependabot` ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the [Security Alerts page](https://github.com/meilisearch/meilisearch/network/alerts).
</details>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
4038: Fix filter escaping issues r=ManyTheFish a=Kerollmops
This PR fixes#4034 by always escaping the sequences. Users must always put quotes (simple or double) to escape the filter values.
Co-authored-by: Kerollmops <clement@meilisearch.com>
3913: Expose a Puffin server to profile the indexing process r=Kerollmops a=Kerollmops
This PR exposes a puffin HTTP server to expose the internal timing it takes to index documents, delete documents, or update the settings of an index.
<img width="1752" alt="Capture d’écran 2023-07-10 à 18 44 58" src="https://github.com/meilisearch/meilisearch/assets/3610253/a3c7a6bf-db5b-42f4-8be1-c4e31c869843">
## To be done
- [x] Move the puffin HTTP server under a feature flag.
- [x] Use [the `puffin::set_scopes_on` function](https://docs.rs/puffin/latest/puffin/fn.set_scopes_on.html) to toggle it (by using the feature directly).
When this function is called with `false`, [a call to `profile_scope!` talked 1-2ns](https://docs.rs/puffin/latest/puffin/fn.set_scopes_on.html).
- [x] Create a _PROFILING.md_ file explaining how to use it.
- [x] Explain that merging scopes on the interface is not always useful.
- [x] Add more info on the number of batched tasks (using the `puffin::profile_scope!` macro data).
- I added more info, but that's more continuous work when we consider we need more info here and there.
- [x] Clean up some scopes, and don't touch too much code to inject puffin.
- I am not sure that the _index_documents/mod.rs_ function is that complex with the addition of the scope.
- [x] Think about what we consider frames. One indexation operation or the wall program. When must we stop the frame, then?
- What we consider a frame is one single `IndexScheduler::tick` execution.
- We can change that later.
Co-authored-by: Kerollmops <clement@meilisearch.com>
Co-authored-by: Clément Renault <clement@meilisearch.com>
3866: Update charabia v0.8.0 r=dureuill a=ManyTheFish
# Pull Request
Update Charabia:
- enhance Japanese segmentation
- enhance Latin Tokenization
- words containing `_` are now properly segmented into several words
- brackets `{([])}` are no more considered as context separators so word separated by brackets are now considered near together for the proximity ranking rule
- fixes#3815
- fixes#3778
- fixes [product#151](https://github.com/meilisearch/product/discussions/151)
> Important note: now the float numbers are segmented around the `.` so `3.22` is segmented as [`3`, `.`, `22`] but the middle dot isn't considered as a hard separator, which means that if we search `3.22` we find documents containing `3.22`
Co-authored-by: ManyTheFish <many@meilisearch.com>
3670: Fix addition deletion bug r=irevoire a=irevoire
The first commit of this PR is a revert of https://github.com/meilisearch/meilisearch/pull/3667. It re-enable the auto-batching of addition and deletion of tasks. No new changes have been introduced outside of `milli`. So all the changes you see on the autobatcher have actually already been reviewed.
It fixes https://github.com/meilisearch/meilisearch/issues/3440.
### What was happening?
The issue was that the `external_documents_ids` generated in the `transform` were used in a very strange way that wasn’t compatible with the deletion of documents.
Instead of doing a clear merge between the external document IDs of the DB and the one returned by the transform + writing it on disk, we were doing some weird tricks with the soft-deleted to avoid writing the fst on disk as much as possible.
The new algorithm may be a bit slower but is way more straightforward and doesn’t change depending on if the soft deletion was used or not. Here is a list of the changes introduced:
1. We now do a clear distinction between the `new_external_documents_ids` coming from the transform and only held on RAM and the `external_documents_ids` coming from the DB.
2. The `new_external_documents_ids` (coming out of the transform) are now represented as an `fst`. We don't need to struggle with the hard, soft distinction + the soft_deleted => That's easier to understand
3. When indexing documents, we merge the `external_documents_ids` coming from the DB and the `new_external_documents_ids` coming from the transform.
### Other things introduced in this PR
Since we constantly have to write small, very specialized fuzzers for this kind of bug, we decided to push the one used to reproduce this bug.
It's not perfect, but it's easy to improve in the future.
It'll also run for as long as possible on every merge on the main branch.
Co-authored-by: Tamo <tamo@meilisearch.com>
Co-authored-by: Loïc Lecrenier <loic.lecrenier@icloud.com>
Conflicts | resolution
----------|-----------
Cargo.lock | added mimalloc
Cargo.toml | took origin/main version
milli/src/search/criteria/exactness.rs | deleted after checking it was only clippy changes
milli/src/search/query_tree.rs | deleted after checking it was only clippy changes
3688: Following release v1.1.1: bring back changes into `main` r=curquiza a=curquiza
`@meilisearch/engine-team` ensure the changes we bring to `main` are the ones you want
Co-authored-by: Louis Dureuil <louis@meilisearch.com>
Co-authored-by: bors[bot] <26634292+bors[bot]@users.noreply.github.com>
Co-authored-by: Tamo <tamo@meilisearch.com>
Co-authored-by: dureuill <dureuill@users.noreply.github.com>
3347: Enhance language detection r=irevoire a=ManyTheFish
## Summary
Some completely unrelated Languages can share the same characters, in Meilisearch we detect the Languages using `whatlang`, which works well on large texts but fails on small search queries leading to a bad segmentation and normalization of the query.
This PR now stores the Languages detected during the indexing in order to reduce the Languages list that can be detected during the search.
## Detail
- Create a 19th database mapping the scripts and the Languages detected with the documents where the Language is detected
- Fill the newly created database during indexing
- Create an allow-list with this database and pass it to Charabia
- Add a test ensuring that a Japanese request containing kanjis only is detected as Japanese and not Chinese
## Related issues
Fixes#2403Fixes#3513
Co-authored-by: f3r10 <frledesma@outlook.com>
Co-authored-by: ManyTheFish <many@meilisearch.com>
Co-authored-by: Many the fish <many@meilisearch.com>
3505: Csv delimiter r=irevoire a=irevoire
Fixes https://github.com/meilisearch/meilisearch/issues/3442
Closes https://github.com/meilisearch/meilisearch/pull/2803
Specified in https://github.com/meilisearch/specifications/pull/221
This PR is a reimplementation of https://github.com/meilisearch/meilisearch/pull/2803, on the new engine. Thanks for your idea and initial PR `@MixusMinimax;` sorry I couldn’t update/merge your PR. Way too many changes happened on the engine in the meantime.
**Attention to reviewer**; I had to update deserr to implement the support of deserializing `char`s
-------
It introduces four new error messages;
- Invalid value in parameter csvDelimiter: expected a string of one character, but found an empty string
- Invalid value in parameter csvDelimiter: expected a string of one character, but found the following string of 5 characters: doggo
- csv delimiter must be an ascii character. Found: 🍰
- The Content-Type application/json does not support the use of a csv delimiter. The csv delimiter can only be used with the Content-Type text/csv.
And one error code;
- `invalid_index_csv_delimiter`
The `invalid_content_type` error code is now also used when we encounter the `csvDelimiter` query parameter with a non-csv content type.
Co-authored-by: Tamo <tamo@meilisearch.com>
3461: Bring v1 changes into main r=curquiza a=Kerollmops
Also bring back changes in milli (the remote repository) into main done during the pre-release
Co-authored-by: Loïc Lecrenier <loic.lecrenier@me.com>
Co-authored-by: bors[bot] <26634292+bors[bot]@users.noreply.github.com>
Co-authored-by: curquiza <curquiza@users.noreply.github.com>
Co-authored-by: Tamo <tamo@meilisearch.com>
Co-authored-by: Philipp Ahlner <philipp@ahlner.com>
Co-authored-by: Kerollmops <clement@meilisearch.com>
3406: Master Key: Implements errors and warnings from the specification r=irevoire a=dureuill
<sub>Now in technicolor</sub>
# Pull Request
## What does this PR do?
- Uses `atty` and `termcolor` as dependency
- Use these dependencies to print colored background for warning messages
- Update messages to match https://github.com/meilisearch/specifications/pull/209
## PR checklist
Please check if your PR fulfills the following requirements:
- [ ] Does this PR fix an existing issue, or have you listed the changes applied in the PR description (and why they are needed)?
- [ ] Have you read the contributing guidelines?
- [ ] Have you made sure that the title is accurate and descriptive of the changes?
Thank you so much for contributing to Meilisearch!
Co-authored-by: Louis Dureuil <louis@meilisearch.com>
