Commit Graph

9564 Commits

Author SHA1 Message Date
Clément Renault
af2b722fed
Make the CI happy about the never type 2024-09-16 11:40:26 +02:00
Clément Renault
8cb7001755
Expose an experimental parameter to control the generation of prefix dbs 2024-09-16 10:57:52 +02:00
meili-bors[bot]
882663bf7f
Merge #4891
4891: Update version for the next release (v1.9.1) in Cargo.toml r=dureuill 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: dureuill <dureuill@users.noreply.github.com>
2024-08-27 16:04:18 +00:00
dureuill
3234f63c00 Update version for the next release (v1.9.1) in Cargo.toml 2024-08-27 16:02:43 +00:00
meili-bors[bot]
9fff081043
Merge #4889
4889: When `retrieveVectors` is true, retrieve `_vectors.embedder` even if … r=Kerollmops a=dureuill

…there are no vector for that embedder


backports a bug fix from v1.10.0: 82647bcded

Co-authored-by: Louis Dureuil <louis@meilisearch.com>
2024-08-27 15:14:58 +00:00
Louis Dureuil
575b7b7a0b
Fix tests 2024-08-27 17:14:10 +02:00
Louis Dureuil
6287f5b204
Remove unexecuted test 2024-08-27 16:54:33 +02:00
Louis Dureuil
5dac8e7168
Allow fuzzing cfg 2024-08-27 16:43:44 +02:00
Louis Dureuil
e669af1e49
CI: Add ACTIONS_ALLOW_USE_UNSECURE_NODE_VERSION workaround to keep using Ubuntu 18.04 2024-08-27 16:29:45 +02:00
Louis Dureuil
0e0e29459c
Update time 2024-08-27 16:27:05 +02:00
Louis Dureuil
c25f7e3450
When retrieveVectors is true, retrieve _vectors.embedder even if there are no vector for that embedder 2024-08-27 16:26:41 +02:00
meili-bors[bot]
0df84bbba7
Merge #4746
4746: Fix hybrid search limit offset r=irevoire a=dureuill

# Pull Request

## Related issue
Fixes #4745

## What does this PR do?
- Apply offset and limit to the keyword search results when they are returned early.
- Add a test that is initially failing, and then passes


Co-authored-by: Louis Dureuil <louis@meilisearch.com>
2024-06-27 12:47:08 +00:00
Louis Dureuil
e53de15b8e
Fix behavior of limit and offset for hybrid search when keyword results are returned early
The test is fixed
2024-06-27 14:25:33 +02:00
Louis Dureuil
8c4921b9dd
Add failing test on limit+offset for hybrid search 2024-06-27 14:21:34 +02:00
meili-bors[bot]
f6a00f4a90
Merge #4740
4740: Make `embeddings` optional and improve error message for `regenerate` r=dureuill a=irevoire

# Pull Request

## Related issue
Fixes https://github.com/meilisearch/meilisearch/issues/4741

## What does this PR do?
- Make the `embeddings` parameter optional when manually specifying embeddings for an embedder
- Adds a lot of tests around malformed `_vectors.embedder` objects
- Use `deserr` to deserialize the `_vectors.embedder` field, improving error messages


Co-authored-by: Tamo <tamo@meilisearch.com>
2024-06-27 10:06:28 +00:00
Tamo
ce08dc509b add more tests and improve the location of the error 2024-06-27 11:51:45 +02:00
Tamo
1daaed163a Make _vectors.:embedding.regenerate mandatory + tests + error messages 2024-06-27 11:04:58 +02:00
meili-bors[bot]
298c7b0c93
Merge #4715
4715: Build all arroy indexes that need to be built r=dureuill a=irevoire

# Pull Request

## Related issue
Fixes https://github.com/meilisearch/meilisearch/issues/4588

## What does this PR do?
- Update arroy
- Ensure we always rebuild the arroy indexes that need to be built


Co-authored-by: Tamo <tamo@meilisearch.com>
2024-06-24 09:32:04 +00:00
Tamo
606e108420 fix all the flaky snapshots 2024-06-24 11:13:45 +02:00
Tamo
7be17b7e4c add the missing snapshots 2024-06-24 10:52:57 +02:00
Tamo
1693332cab Update arroy and always build the tree that need to be built 2024-06-24 10:14:03 +02:00
meili-bors[bot]
ddd564665b
Merge #4713
4713: Speed up facet distribution r=ManyTheFish a=Kerollmops

This PR is akin to #4682, but this time, the same logic is applied to the facets. Bitmaps are not decoded, and we do an intersection on the bytes with the search candidates instead of materializing the RoaringBitmap to destroy it just after the operation.

A prospect raised some slow requests when performing facet searches, and I found out that the disk optimization intersection wasn't performed on the facets.

Co-authored-by: Clément Renault <clement@meilisearch.com>
2024-06-24 05:23:46 +00:00
meili-bors[bot]
4ae11bfd31
Merge #4710
4710: Only spawn thread pool once (v1.9) r=irevoire a=dureuill

# Pull Request

See #4707 

Co-authored-by: Louis Dureuil <louis@meilisearch.com>
2024-06-20 11:45:32 +00:00
Clément Renault
9736e16a88
Make clippy happy 2024-06-20 13:02:44 +02:00
Clément Renault
6fa4da8ae7
Improve facet distribution speed in count mode 2024-06-20 12:58:51 +02:00
Clément Renault
19d7cdc20d
Improve facet distribution speed in lexico mode 2024-06-20 12:57:08 +02:00
meili-bors[bot]
c229200820
Merge #4712
4712: Update mini-dashboard 2.14 r=irevoire a=curquiza

Fixes #4668

Co-authored-by: curquiza <clementine@meilisearch.com>
2024-06-20 08:47:22 +00:00
curquiza
bad28cc9e2 Update mini-dashboard 2.14 2024-06-20 10:01:36 +02:00
Louis Dureuil
a04041c8f2
Only spawn the pool once 2024-06-19 16:25:33 +02:00
meili-bors[bot]
e580d6b98f
Merge #4693
4693: Introduce distinct attributes at search time r=irevoire a=Kerollmops

This PR fixes #4611.

### To Do
- [x] Remove the `distinguishableAttributes` settings (not even a commit about that).
- [x] Use the `filterableAttributes` to be able to use the `distinct` parameter at search.
- [x] Work on the errors and make tests.

Co-authored-by: Clément Renault <clement@meilisearch.com>
Co-authored-by: Tamo <tamo@meilisearch.com>
2024-06-18 07:45:03 +00:00
Tamo
8ba65e333b add snapshot files 2024-06-17 16:50:26 +02:00
Tamo
43875e6758 fix bug around nested fields 2024-06-17 15:59:30 +02:00
Tamo
d7844a6e45 add a bunch of tests on the errors of the distinct at search time 2024-06-17 15:37:32 +02:00
meili-bors[bot]
e9bf4c43a4
Merge #4649
4649: Don't store the vectors in the documents database r=dureuill a=irevoire

# Pull Request

## Related issue
Fixes https://github.com/meilisearch/meilisearch/issues/4607

## What does this PR do?
- Ensure that anything falling under `_vectors` is NOT searchable, filterable or sortable
- [x] per embedder, add a roaring bitmap of documents that provide "userProvided" embeddings
- [x] in the indexing process in extract_vector_points, set the bit corresponding to the document depending on the "userProvided" subfield in the _vectors field.
- [x] in the document DB in typed chunks, when writing the _vectors field, remove all keys corresponding to an embedder

Co-authored-by: Tamo <tamo@meilisearch.com>
Co-authored-by: Louis Dureuil <louis@meilisearch.com>
2024-06-17 12:32:03 +00:00
Tamo
a8a0854421
Update meilisearch/src/analytics/segment_analytics.rs 2024-06-17 14:30:50 +02:00
Louis Dureuil
0a8f50695e
Fixes for Rust v1.79 2024-06-13 17:47:44 +02:00
Louis Dureuil
09d9b63e1c
- test case where all vectors were generated
- update tests following changes in behavior from previous commit
2024-06-13 17:16:41 +02:00
Louis Dureuil
b9b938c902
Change retrieveVectors behavior:
- when the feature is disabled, documents are never modified
- when the feature is enabled and `retrieveVectors` is disabled, `_vectors` is removed from documents
- when the feature is enabled and `retrieveVectors` is enabled, vectors from the vectors DB are merged with `_vectors` in documents

Additionally `_vectors` is never displayed when the `displayedAttributes` list does not contain either `*` or `_vectors`

- fixed an issue where `_vectors` was not injected when all vectors in the dataset where always generated
2024-06-13 17:13:36 +02:00
Tamo
6bf07d969e add failing test 2024-06-13 15:49:42 +02:00
Louis Dureuil
e35ef31738
Small changes following review 2024-06-13 14:20:48 +02:00
Louis Dureuil
3f212a8202
Update tests 2024-06-12 18:13:34 +02:00
Louis Dureuil
bc547dad6f
Update dump file 2024-06-12 18:12:56 +02:00
Louis Dureuil
3bc8f81abc
user_provided => regenerate 2024-06-12 18:12:20 +02:00
Louis Dureuil
a89eea233b
Fix vectors injection 2024-06-12 17:10:19 +02:00
Louis Dureuil
34fabed214
Add test for vector writeback 2024-06-12 17:09:34 +02:00
Louis Dureuil
fca9fe39b3
Update test snapshots 2024-06-12 14:50:55 +02:00
Louis Dureuil
f5cf01e7d1
Rework extraction to use EmbedderAction 2024-06-12 14:50:55 +02:00
Louis Dureuil
d1dd7e5d09
In transform for removed embedders, write back their user provided vectors in documents, and clear the writers 2024-06-12 14:50:55 +02:00
Louis Dureuil
d18c1f77d7
Update embedder configs with a finer granularity
- no longer clear vector DB between any two embedder changes
2024-06-12 14:50:55 +02:00
Louis Dureuil
d0b05ae691
Add EmbedderAction to settings 2024-06-12 14:50:54 +02:00