4512: Revert "Revert "Merge remote-tracking branch 'origin/main' into release-v1.7.1"" r=Kerollmops a=irevoire
Reverts meilisearch/meilisearch#4510
This PR was supposed to be merged on `release-v1.7.1` not main 🤦
Co-authored-by: Tamo <irevoire@protonmail.ch>
handles the niche case 🐩 in the hybrid search where:
1. a sort ranking rule is the first rule.
2. the keyword search is skipped at the first rule.
3. the semantic search is not skipped at the first rule.
Previously, we would have the skipped search winning, whereas we want the non skipped one winning.
4510: Revert "Merge remote-tracking branch 'origin/main' into release-v1.7.1" r=Kerollmops a=irevoire
In https://github.com/meilisearch/meilisearch/pull/4502 we merged main into release-v1.7.1 instead of a temporary branch thus we now need to revert this merge commit.
This reverts commit bd74cce86a5e610477cc68bd555a636bcd735ca3, reversing changes made to d2f77e88bd9f9ce018a71e78c120274336e80580.
Co-authored-by: Tamo <tamo@meilisearch.com>
4500: Don't display dimensions as 0 when it is not set r=ManyTheFish a=dureuill
Fixes regression in embedders where `dimensions: 0` was displayed when it hadn't be set for the `openAi` source.
Was breaking a PHP SDK integration test: cbaecb8c55/tests/Settings/EmbeddersTest.php (L28)
Co-authored-by: Louis Dureuil <louis@meilisearch.com>
4499: Fix milli link in contributing doc r=curquiza a=mohsen-alizadeh
# Pull Request
## Related issue
Fixes#4498
## What does this PR do?
The milli link in CONTRIBUTING.md targeted the archived milli repository. it has to be changed to target to the milli crate in the main repo
## PR checklist
Please check if your PR fulfills the following requirements:
- [X] Does this PR fix an existing issue, or have you listed the changes applied in the PR description (and why they are needed)?
- [X] Have you read the contributing guidelines?
- [X] Have you made sure that the title is accurate and descriptive of the changes?
Thank you so much for contributing to Meilisearch!
Co-authored-by: Mohsen Alizadeh <mohsen@alizadeh.us>
Co-authored-by: Clémentine U. - curqui <clementine@meilisearch.com>
4491: chore: remove repetitive words r=curquiza a=shuangcui
# Pull Request
## Related issue
Fixes #<issue_number>
## What does this PR do?
- ...
## 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: shuangcui <fliter@qq.com>
4483: Workflows: Fix reason param when benches are triggered from a comment. r=irevoire a=dureuill
Co-authored-by: Louis Dureuil <louis@meilisearch.com>
4479: Skip reindexing when modifying unknown faceted fields r=dureuill a=Kerollmops
This PR improves Meilisearch's decision to reindex when a faceted field is added to the settings, but not a single document contains this field. It is effectively a waste of time to reindex documents when the engine needs to know a field.
This is related to a conversation [we have with our biggest customer (internal link)](https://discord.com/channels/1006923006964154428/1101213808627830794/1217112918857089187). They have 170 million documents, so reindexing this amount would be problematic.
---
The image is available by using the following Docker command. You can see the advancement of the image's build [on the GitHub CI page](https://github.com/meilisearch/meilisearch/actions/runs/8251688778).
```
docker pull getmeili/meilisearch:prototype-no-reindex-unknown-fields-0
```
Here is the hand-made test that shows that when modifying unknown filterable attributes, here `lol`, it doesn't reindex. However, when modifying the known `genre` field, it does reindex. You can see all that by looking at the time spent processing the update.
```json
{
"uid": 3,
"indexUid": "movies",
"status": "succeeded",
"type": "settingsUpdate",
"canceledBy": null,
"details": {
"filterableAttributes": [
"genres"
]
},
"error": null,
"duration": "PT9.237703S",
"enqueuedAt": "2024-03-12T15:34:26.836083Z",
"startedAt": "2024-03-12T15:34:26.836374Z",
"finishedAt": "2024-03-12T15:34:36.074077Z"
},
{
"uid": 2,
"indexUid": "movies",
"status": "succeeded",
"type": "settingsUpdate",
"canceledBy": null,
"details": {
"filterableAttributes": [
"lol"
]
},
"error": null,
"duration": "PT0.000751S",
"enqueuedAt": "2024-03-12T15:33:53.563923Z",
"startedAt": "2024-03-12T15:33:53.565259Z",
"finishedAt": "2024-03-12T15:33:53.56601Z"
},
{
"uid": 0,
"indexUid": "movies",
"status": "succeeded",
"type": "documentAdditionOrUpdate",
"canceledBy": null,
"details": {
"receivedDocuments": 31944,
"indexedDocuments": 31944
},
"error": null,
"duration": "PT3.120723S",
"enqueuedAt": "2024-02-17T10:35:55.042864Z",
"startedAt": "2024-02-17T10:35:55.043505Z",
"finishedAt": "2024-02-17T10:35:58.164228Z"
}
```
Co-authored-by: Clément Renault <clement@meilisearch.com>
4487: Update version for the next release (v1.7.1) in Cargo.toml r=Kerollmops a=meili-bot
⚠️ This PR is automatically generated. Check the new version is the expected one and Cargo.lock has been updated before merging.
Co-authored-by: Kerollmops <Kerollmops@users.noreply.github.com>
4476: Make the `/facet-search` route use the `sortFacetValuesBy` setting r=irevoire a=Kerollmops
This PR fixes#4423 by ensuring that the `/facet-search` route uses the `sortFacetValuesBy` setting.
Note for the documentation team (to be moved in the tracking issue): Using the new `sortFacetValuesBy` setting can slow down the facet-search requests as Meilisearch iterates over the whole list of facet values and computes the count of documents on every entry. That is hardly or even impossible to optimize correctly.
### TODO
- [x] Create a custom HashMap wrapper for the facet `OrderBy` settings.
This wrapper will return the `OrderBy` setting of the facet, if not defined will use the default `*` one, and if not there either (strange) will fall back on the lexicographic one.
- [x] Create a `ValuesCollection` wrapper that implements the logic for the lexicographic and count order by.
- [x] Use it when there is no search query.
- [x] Use it when there is a search query with and without allowed typos.
- [x] Do not change the original logic, only use a wrapper.
- [x] Add tests
Co-authored-by: Clément Renault <clement@meilisearch.com>
4475: Allow running benchmarks without sending results to the dashboard r=irevoire a=dureuill
Adds a `--no-dashboard` option to avoid sending results to the dashboard.
Co-authored-by: Louis Dureuil <louis@meilisearch.com>
4456: Add Ollama as an embeddings provider r=dureuill a=jakobklemm
# Pull Request
## Related issue
[Related Discord Thread](https://discord.com/channels/1006923006964154428/1211977150316683305)
## What does this PR do?
- Adds Ollama as a provider of Embeddings besides HuggingFace and OpenAI under the name `ollama`
- Adds the environment variable `MEILI_OLLAMA_URL` to set the embeddings URL of an Ollama instance with a default value of `http://localhost:11434/api/embeddings` if no variable is set
- Changes some of the structs and functions in `openai.rs` to be public so that they can be shared.
- Added more error variants for Ollama specific errors
- It uses the model `nomic-embed-text` as default, but any string value is allowed, however it won't automatically check if the model actually exists or is an embedding model
Tested against Ollama version `v0.1.27` and the `nomic-embed-text` model.
## PR checklist
Please check if your PR fulfills the following requirements:
- [x] Does this PR fix an existing issue, or have you listed the changes applied in the PR description (and why they are needed)?
- [x] Have you read the contributing guidelines?
- [x] Have you made sure that the title is accurate and descriptive of the changes?
Co-authored-by: Jakob Klemm <jakob@jeykey.net>
Co-authored-by: Louis Dureuil <louis.dureuil@gmail.com>
Instead of the user manually specifying the model dimensions it will now automatically get determined
Just like with hf.rs the word "test" gets embedded to determine the dimensions of the output
Add a dedicated error type for if the model doesn't exist (don't automatically pull it though) and set the fault of that error to be the user
