2357: chore(dump): add dump tests r=Kerollmops a=irevoire
Add tests on the import of dump v1, v2, v3 and v4.
Since the dumps are slow to decompress, I made the `flate2` crate always compile in optimized.
And since they're also slow to index, I also made the `milli` crate always compile in optimized. What do you think of this `@MarinPostma?`
Should we keep milli unoptimized in case it could help us debug some things? 👀
Co-authored-by: Tamo <tamo@meilisearch.com>
2316: Add version flag r=Kerollmops a=sanders41
# Pull Request
## What does this PR do?
Fixes#2315
## PR checklist
Please check if your PR fulfills the following requirements:
- [x] Does this PR fix an existing issue?
- [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: Paul Sanders <psanders1@gmail.com>
2313: fix(search): remove the back and forth between the IndexMap and the serde_json::Map r=irevoire a=irevoire
This is ok because we're using the preserve_order feature in serde_json which is already internally using an IndexMap.
See https://github.com/meilisearch/meilisearch/pull/2298#discussion_r845228412_
Co-authored-by: Tamo <tamo@meilisearch.com>
2297: Feat(Search): Enhance formating search results r=ManyTheFish a=ManyTheFish
Add new settings and change crop_len behavior to count words instead of characters.
- [x] `highlightPreTag`
- [x] `highlightPostTag`
- [x] `cropMarker`
- [x] `cropLength` count word instead of chars
- [x] `cropLength` 0 is now considered as no `cropLength`
- [ ] ~smart crop finding the best matches interval~ (postponed)
Partially fixes #2214. (no smart crop)
Co-authored-by: ManyTheFish <many@meilisearch.com>
2277: fix(http): fix panic when sending document update without content type header r=MarinPostma a=MarinPostma
I found a panic when pushing documents without a content-type. This fixes is by returning unknown instead of crashing.
Co-authored-by: ad hoc <postma.marin@protonmail.com>
2207: Fix: avoid embedding the user input into the error response. r=Kerollmops a=CNLHC
# Pull Request
## What does this PR do?
Fix#2107.
The problem is meilisearch embeds the user input to the error message.
The reason for this problem is `milli` throws a `serde_json: Error` whose `Display` implementation will do this embedding.
I tried to solve this problem in this PR by manually implementing the `Display` trait for `DocumentFormatError` instead of deriving automatically.
<!-- Please link the issue you're trying to fix with this PR, if none then please create an issue first. -->
## PR checklist
Please check if your PR fulfills the following requirements:
- [x] Does this PR fix an existing issue?
- [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: Liu Hancheng <cn_lhc@qq.com>
Co-authored-by: LiuHanCheng <2463765697@qq.com>
2281: Hard limit the number of results returned by a search r=Kerollmops a=Kerollmops
This PR fixes#2133 by hard-limiting the number of results that a search request can return at any time. I would like the guidance of `@MarinPostma` to test that, should I use a mocking test here? Or should I do anything else?
I talked about touching the _nb_hits_ value with `@qdequele` and we concluded that it was not correct to do so.
Could you please confirm that it is the right place to change that?
Co-authored-by: Kerollmops <clement@meilisearch.com>
2253: refactor authentication key extraction r=ManyTheFish a=MarinPostma
I am concerned that the part of the code that performs the key prefix extraction from the jwt token migh be misused in the future. Since this is a critical part of the code, I moved it into it's own function. Since we deserialized the payload twice anyway, I reordered the verifications, and we now use the data from the validated token.
Co-authored-by: ad hoc <postma.marin@protonmail.com>
