Compare commits

...

77 Commits

Author SHA1 Message Date
Clément Renault
6c1739218c Settings in queue 2025-05-21 18:34:11 +02:00
Clément Renault
72d4998dce Correctly list the chat settings key actions 2025-05-21 16:24:51 +02:00
Clément Renault
fde11573da Always use the frequency matching strategy 2025-05-21 16:18:37 +02:00
Clément Renault
41220f786b Remove templating validation 2025-05-21 16:10:31 +02:00
Clément Renault
4d59fdb65d Correctly support document templates on the chat API 2025-05-21 15:32:34 +02:00
Clément Renault
3e51c0a4c1 Introduce the new index chat settings 2025-05-21 11:07:06 +02:00
Clément Renault
91c6ab8392 Make sure errorneous calls are handled and forwarded to the LLM 2025-05-20 18:01:08 +02:00
Clément Renault
beff6adeb1 Catch invalid argument calls to search function 2025-05-20 17:55:21 +02:00
Clément Renault
18eab165a7 Support multiple indexes and not only main 2025-05-20 17:43:24 +02:00
Clément Renault
5c6b63df65 Limit the number of internal loop calls and change the function name 2025-05-20 16:44:28 +02:00
Clément Renault
7266aed770 Correctly support tenant tokens and filters 2025-05-20 16:15:49 +02:00
Clément Renault
bae6c98aa3 Stream errors 2025-05-20 12:23:22 +02:00
Clément Renault
42c95cf3c4 Stop the stream when the connexion stops and chnage the events 2025-05-20 12:05:51 +02:00
Clément Renault
4f919db344 Generate a new default chat API key 2025-05-20 11:00:19 +02:00
Clément Renault
295840d07a Change the /chat route to /chat/completions to be OpenAI-compatible 2025-05-20 10:14:56 +02:00
Clément Renault
c0c3bddda8 Better stop the stream 2025-05-16 17:12:48 +02:00
Clément Renault
10b5fcd4ba Update the streaming detection to work with Mistral 2025-05-16 15:17:01 +02:00
Clément Renault
8113d4a52e Make it compatible with the Mistral API 2025-05-16 14:33:53 +02:00
Clément Renault
5964289284 Support base_api in the settings 2025-05-15 18:28:02 +02:00
Clément Renault
6b81854d48 Make clippy happy 2025-05-15 18:16:06 +02:00
Clément Renault
9e5b466426 Display pre-query prompt in search tool response 2025-05-15 18:10:09 +02:00
Clément Renault
b43ffd8fac Commit when putting stuff in LMDB 2025-05-15 18:03:26 +02:00
Clément Renault
43da2bcb8c Remove useless function 2025-05-15 17:52:26 +02:00
Clément Renault
5e3b126d73 Expose new chat settings routes 2025-05-15 17:48:10 +02:00
Clément Renault
6c034754ca Factorise a bit the code 2025-05-15 15:39:38 +02:00
Clément Renault
6329cf7ed6 Display the different tool calls we need to do 2025-05-15 11:17:34 +02:00
Clément Renault
e0c8c11a94 Send an event with the content of the tool calling 2025-05-14 17:15:32 +02:00
Clément Renault
6e8b371111 Streaming supports tool calling 2025-05-14 14:58:01 +02:00
Clément Renault
da7d651f4b Nearly support tools on the streaming route 2025-05-14 14:29:41 +02:00
Clément Renault
24050f06e4 Return the right message format 2025-05-14 12:03:43 +02:00
Clément Renault
af482d8ee9 Aggregate tool calls and display the calls to make. 2025-05-14 11:53:03 +02:00
Clément Renault
7d62307739 Implement a first version of a streamed chat API 2025-05-14 11:18:21 +02:00
Clément Renault
3a71df7b5a Make it work by retrieving content from the index 2025-05-13 16:35:46 +02:00
Clément Renault
ac39a436d9 Support overwriten prompts of the search query 2025-05-13 16:33:58 +02:00
Clément Renault
e5c963a170 Support querying the index named main 2025-05-13 15:26:24 +02:00
Clément Renault
9baf2ce1a6 Introduce the first version of the /chat route that mimics the OpenAI API 2025-05-13 11:19:32 +02:00
Many the fish
d9a527854a Merge pull request #5546 from meilisearch/curquiza-patch-1
Add set in GitHub action to notify deployment to Swarmia
2025-05-12 13:36:13 +00:00
Louis Dureuil
e4f05326be Merge pull request #5552 from meilisearch/v1-15-dumpless-upgrade
Add v1.15 in index-scheduler upgrade
2025-05-12 12:59:47 +00:00
ManyTheFish
d99419acfb Add a NoOp operation in index update 2025-05-12 14:19:15 +02:00
ManyTheFish
f349630e78 Add v1.15 in index-scheduler upgrade 2025-05-12 13:53:23 +02:00
Many the fish
2d1412afce Merge pull request #5549 from meilisearch/update-version-v1.15.0
Update version for the next release (v1.15.0) in Cargo.toml
2025-05-12 09:21:43 +00:00
ManyTheFish
0f4536df2d Adapt dumpless upgrade tests 2025-05-12 10:43:12 +02:00
ManyTheFish
3531efb169 Update version for the next release (v1.15.0) in Cargo.toml 2025-05-12 08:04:18 +00:00
Clémentine
6ec430b633 Update .github/workflows/publish-docker-images.yml 2025-05-08 20:08:34 +02:00
Clémentine
4041978402 Add set in GitHub action to notify deployment to Swarmia 2025-05-08 20:07:36 +02:00
Many the fish
71ab11f1fe Merge pull request #5523 from meilisearch/rollback-updates
Allow rollbacking updates
2025-05-05 09:53:56 +00:00
Many the fish
436776cdbf Merge pull request #5535 from meilisearch/filter-comparison-string
Allow lexicographic filtering of strings
2025-05-05 09:53:19 +00:00
Many the fish
96bc519f9e Merge pull request #5494 from meilisearch/deactivate-numbers-in-typos
Deactivate numbers in typos
2025-05-05 09:19:53 +00:00
ManyTheFish
8b23eddc10 Dumpless upgrade 2025-04-30 18:03:50 +02:00
Louis Dureuil
185f2b8f74 Fix test now that lexicographic string comparisons are allowed 2025-04-30 17:28:59 +02:00
Louis Dureuil
c0e987979a Allow lexicographic string comparisons 2025-04-30 17:28:49 +02:00
Clément Renault
032c67662d Merge pull request #5533 from ZeroZ-lab/fix-readme
Fix links and formatting in README.md for clarity and consistency
2025-04-29 22:11:36 +00:00
GuanMu
03f59786c2 Fix links and formatting in README.md for clarity and consistency 2025-04-30 00:10:41 +08:00
Louis Dureuil
f7c1f19dd8 rust fmt 2025-04-29 16:10:43 +02:00
Louis Dureuil
1542ff30ae Roll back index scheduler version first 2025-04-29 16:05:43 +02:00
Louis Dureuil
20d0aa499a Apply suggestions from code review
Co-authored-by: Clément Renault <clement@meilisearch.com>
2025-04-29 16:03:30 +02:00
ManyTheFish
0cb2bf34a5 Fix test 2025-04-29 14:47:30 +02:00
Louis Dureuil
de03b7e437 Merge pull request #5530 from meilisearch/rename-batcher-stopped-because
Rename `batcherStoppedBecause` to `batchCreationComplete`
2025-04-29 10:35:57 +00:00
Louis Dureuil
a315726f96 Update snapshots 2025-04-29 11:50:32 +02:00
Louis Dureuil
91d2a07499 Rename batcherStoppedBecause to batchCreationComplete 2025-04-29 10:40:12 +02:00
Louis Dureuil
c5360bcdbf When canceling an upgrade task, execute the rollback code 2025-04-24 16:59:03 +02:00
Louis Dureuil
1bdc08a73a tick: always refuse to batch tasks when the version in the index-scheduler is wrong 2025-04-24 16:54:43 +02:00
Louis Dureuil
63b5e21ae1 tick: check tasks to cancel before checking for upgrade tasks 2025-04-24 16:52:28 +02:00
Louis Dureuil
eb0b5239cb process rollback 2025-04-24 16:52:28 +02:00
Louis Dureuil
121c1ac1dd Upgrade supports cancelling 2025-04-24 16:08:10 +02:00
Louis Dureuil
b82dda2d0d Allow rollbacking indexes in the mapper 2025-04-24 16:08:10 +02:00
Louis Dureuil
ea9330e9c9 Add new errors when there is a version mismatch between the bin and index or index-scheduler 2025-04-24 16:08:10 +02:00
Louis Dureuil
b6a9d8d2ac Add Error::RollbackFailed 2025-04-24 16:06:19 +02:00
Louis Dureuil
a03eef6511 Support rollback 2025-04-24 16:06:19 +02:00
Louis Dureuil
42fae9994d Move tests out of index.rs 2025-04-24 16:06:19 +02:00
Louis Dureuil
e1aa534389 Wait 10 seconds in case of irrecoverable error 2025-04-24 16:06:19 +02:00
Louis Dureuil
49add50cb3 Make version constants u32 2025-04-24 16:06:19 +02:00
Louis Dureuil
29b947ee43 make Index::get_version public 2025-04-24 16:05:52 +02:00
Clément Renault
3f683c4238 Merge pull request #5525 from meilisearch/arroy-call-tracking
Display the time spent querying the vector store
2025-04-23 20:37:26 +00:00
ManyTheFish
294ccb6f44 Add test 2025-04-23 16:57:50 +02:00
ManyTheFish
63a4dfa2a8 Add disableOnNumber setting 2025-04-23 16:57:50 +02:00
Kerollmops
3b8965bc76 Display and sum the time spent in arroy 2025-04-22 18:10:42 +02:00
96 changed files with 4350 additions and 1871 deletions

View File

@@ -104,3 +104,20 @@ jobs:
repository: meilisearch/meilisearch-cloud
event-type: cloud-docker-build
client-payload: '{ "meilisearch_version": "${{ github.ref_name }}", "stable": "${{ steps.check-tag-format.outputs.stable }}" }'
# Send notification to Swarmia to notify of a deployment: https://app.swarmia.com
- name: Send deployment to Swarmia
if: github.event_name == 'push' && success()
run: |
JSON_STRING=$( jq --null-input --compact-output \
--arg version "${{ github.ref_name }}" \
--arg appName "meilisearch" \
--arg environment "production" \
--arg commitSha "${{ github.sha }}" \
--arg repositoryFullName "${{ github.repository }}" \
'{"version": $version, "appName": $appName, "environment": $environment, "commitSha": $commitSha, "repositoryFullName": $repositoryFullName}' )
curl -H "Authorization: ${{ secrets.SWARMIA_DEPLOYMENTS_AUTHORIZATION }}" \
-H "Content-Type: application/json" \
-d "$JSON_STRING" \
https://hook.swarmia.com/deployments

594
Cargo.lock generated

File diff suppressed because it is too large Load Diff

View File

@@ -22,7 +22,7 @@ members = [
]
[workspace.package]
version = "1.14.0"
version = "1.15.0"
authors = [
"Quentin de Quelen <quentin@dequelen.me>",
"Clément Renault <clement@meilisearch.com>",

View File

@@ -41,7 +41,7 @@
- [**Movies**](https://where2watch.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=organization) — An application to help you find streaming platforms to watch movies using [hybrid search](https://www.meilisearch.com/solutions/hybrid-search?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demos).
- [**Ecommerce**](https://ecommerce.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demos) — Ecommerce website using disjunctive [facets](https://www.meilisearch.com/docs/learn/fine_tuning_results/faceted_search?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demos), range and rating filtering, and pagination.
- [**Songs**](https://music.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demos) — Search through 47 million of songs.
- [**SaaS**](https://saas.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demos) — Search for contacts, deals, and companies in this [multi-tenant](https://www.meilisearch.com/docs/learn/security/multitenancy_tenant_tokens?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demos) CRM application.
- [**SaaS**](https://saas.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demos) — Search for contacts, deals, and companies in this [multi-tenant](https://www.meilisearch.com/docs/learn/security/multitenancy_tenant_tokens?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=demos) CRM application.
See the list of all our example apps in our [demos repository](https://github.com/meilisearch/demos).
@@ -99,7 +99,7 @@ If you want to know more about the kind of data we collect and what we use it fo
## 📫 Get in touch!
Meilisearch is a search engine created by [Meili]([https://www.welcometothejungle.com/en/companies/meilisearch](https://www.meilisearch.com/careers)), a software development company headquartered in France and with team members all over the world. Want to know more about us? [Check out our blog!](https://blog.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=contact)
Meilisearch is a search engine created by [Meili](https://www.meilisearch.com/careers), a software development company headquartered in France and with team members all over the world. Want to know more about us? [Check out our blog!](https://blog.meilisearch.com/?utm_campaign=oss&utm_source=github&utm_medium=meilisearch&utm_content=contact)
🗞 [Subscribe to our newsletter](https://meilisearch.us2.list-manage.com/subscribe?u=27870f7b71c908a8b359599fb&id=79582d828e) if you don't want to miss any updates! We promise we won't clutter your mailbox: we only send one edition every two months.

View File

@@ -31,7 +31,7 @@ anyhow = "1.0.95"
bytes = "1.9.0"
convert_case = "0.6.0"
flate2 = "1.0.35"
reqwest = { version = "0.12.12", features = ["blocking", "rustls-tls"], default-features = false }
reqwest = { version = "0.12.15", features = ["blocking", "rustls-tls"], default-features = false }
[features]
default = ["milli/all-tokenizations"]

View File

@@ -373,6 +373,7 @@ impl<T> From<v5::Settings<T>> for v6::Settings<v6::Unchecked> {
},
disable_on_words: typo.disable_on_words.into(),
disable_on_attributes: typo.disable_on_attributes.into(),
disable_on_numbers: v6::Setting::NotSet,
}),
v5::Setting::Reset => v6::Setting::Reset,
v5::Setting::NotSet => v6::Setting::NotSet,
@@ -397,6 +398,7 @@ impl<T> From<v5::Settings<T>> for v6::Settings<v6::Unchecked> {
search_cutoff_ms: v6::Setting::NotSet,
facet_search: v6::Setting::NotSet,
prefix_search: v6::Setting::NotSet,
chat: v6::Setting::NotSet,
_kind: std::marker::PhantomData,
}
}

View File

@@ -2,6 +2,7 @@ use std::fmt::Display;
use meilisearch_types::batches::BatchId;
use meilisearch_types::error::{Code, ErrorCode};
use meilisearch_types::milli::index::RollbackOutcome;
use meilisearch_types::tasks::{Kind, Status};
use meilisearch_types::{heed, milli};
use thiserror::Error;
@@ -150,8 +151,24 @@ pub enum Error {
CorruptedTaskQueue,
#[error(transparent)]
DatabaseUpgrade(Box<Self>),
#[error("Failed to rollback for index `{index}`: {rollback_outcome} ")]
RollbackFailed { index: String, rollback_outcome: RollbackOutcome },
#[error(transparent)]
UnrecoverableError(Box<Self>),
#[error("The index scheduler is in version v{}.{}.{}, but Meilisearch is in version v{}.{}.{}.\n - hint: start the correct version of Meilisearch, or consider updating your database. See also <https://www.meilisearch.com/docs/learn/update_and_migration/updating>",
index_scheduler_version.0, index_scheduler_version.1, index_scheduler_version.2,
package_version.0, package_version.1, package_version.2)]
IndexSchedulerVersionMismatch {
index_scheduler_version: (u32, u32, u32),
package_version: (u32, u32, u32),
},
#[error("Index `{index}` is in version v{}.{}.{}, but Meilisearch is in version v{}.{}.{}.\n - note: this is an internal error, please consider filing a bug report: <https://github.com/meilisearch/meilisearch/issues/new?template=bug_report.md>",
index_version.0, index_version.1, index_version.2, package_version.0, package_version.1, package_version.2)]
IndexVersionMismatch {
index: String,
index_version: (u32, u32, u32),
package_version: (u32, u32, u32),
},
#[error(transparent)]
HeedTransaction(heed::Error),
@@ -209,6 +226,9 @@ impl Error {
| Error::CorruptedTaskQueue
| Error::DatabaseUpgrade(_)
| Error::UnrecoverableError(_)
| Error::IndexSchedulerVersionMismatch { .. }
| Error::IndexVersionMismatch { .. }
| Error::RollbackFailed { .. }
| Error::HeedTransaction(_) => false,
#[cfg(test)]
Error::PlannedFailure => false,
@@ -274,7 +294,10 @@ impl ErrorCode for Error {
Error::CorruptedTaskQueue => Code::Internal,
Error::CorruptedDump => Code::Internal,
Error::DatabaseUpgrade(_) => Code::Internal,
Error::RollbackFailed { .. } => Code::Internal,
Error::UnrecoverableError(_) => Code::Internal,
Error::IndexSchedulerVersionMismatch { .. } => Code::Internal,
Error::IndexVersionMismatch { .. } => Code::Internal,
Error::CreateBatch(_) => Code::Internal,
// This one should never be seen by the end user

View File

@@ -7,6 +7,7 @@ use meilisearch_types::heed::types::{SerdeJson, Str};
use meilisearch_types::heed::{Database, Env, RoTxn, RwTxn, WithoutTls};
use meilisearch_types::milli;
use meilisearch_types::milli::database_stats::DatabaseStats;
use meilisearch_types::milli::index::RollbackOutcome;
use meilisearch_types::milli::update::IndexerConfig;
use meilisearch_types::milli::{FieldDistribution, Index};
use serde::{Deserialize, Serialize};
@@ -431,6 +432,51 @@ impl IndexMapper {
Ok(index)
}
pub fn rollback_index(
&self,
rtxn: &RoTxn,
name: &str,
to: (u32, u32, u32),
) -> Result<RollbackOutcome> {
// remove any currently updating index to make sure that we aren't keeping a reference to the index somewhere
drop(self.currently_updating_index.write().unwrap().take());
let uuid = self
.index_mapping
.get(rtxn, name)?
.ok_or_else(|| Error::IndexNotFound(name.to_string()))?;
// take the lock to make sure noone is messing with the indexes while we rollback
// this will block any search or other operation, but we are rollbacking so this is probably acceptable.
let mut index_map = self.index_map.write().unwrap();
'close_index: loop {
match index_map.get(&uuid) {
Available(_) => {
index_map.close_for_resize(&uuid, self.enable_mdb_writemap, 0);
// index should now be `Closing`; try again
continue;
}
// index already closed
Missing => break 'close_index,
// closing requested by this thread or another one; wait for closing to complete, then exit
Closing(closing_index) => {
if closing_index.wait_timeout(Duration::from_secs(100)).is_none() {
// release the lock so it doesn't get poisoned
drop(index_map);
panic!("cannot close index")
}
break;
}
BeingDeleted => return Err(Error::IndexNotFound(name.to_string())),
};
}
let index_path = self.base_path.join(uuid.to_string());
Index::rollback(milli::heed::EnvOpenOptions::new().read_txn_without_tls(), index_path, to)
.map_err(|err| crate::Error::from_milli(err, Some(name.to_string())))
}
/// Attempts `f` for each index that exists in the index mapper.
///
/// It is preferable to use this function rather than a loop that opens all indexes, as a way to avoid having all indexes opened,

View File

@@ -41,11 +41,8 @@ pub fn snapshot_index_scheduler(scheduler: &IndexScheduler) -> String {
let mut snap = String::new();
let indx_sched_version = version.get_version(&rtxn).unwrap();
let latest_version = (
versioning::VERSION_MAJOR.parse().unwrap(),
versioning::VERSION_MINOR.parse().unwrap(),
versioning::VERSION_PATCH.parse().unwrap(),
);
let latest_version =
(versioning::VERSION_MAJOR, versioning::VERSION_MINOR, versioning::VERSION_PATCH);
if indx_sched_version != Some(latest_version) {
snap.push_str(&format!("index scheduler running on version {indx_sched_version:?}\n"));
}

View File

@@ -28,6 +28,7 @@ mod lru;
mod processing;
mod queue;
mod scheduler;
mod settings;
#[cfg(test)]
mod test_utils;
pub mod upgrade;
@@ -53,8 +54,8 @@ use flate2::Compression;
use meilisearch_types::batches::Batch;
use meilisearch_types::features::{InstanceTogglableFeatures, Network, RuntimeTogglableFeatures};
use meilisearch_types::heed::byteorder::BE;
use meilisearch_types::heed::types::I128;
use meilisearch_types::heed::{self, Env, RoTxn, WithoutTls};
use meilisearch_types::heed::types::{SerdeJson, Str, I128};
use meilisearch_types::heed::{self, Database, Env, RoTxn, Unspecified, WithoutTls};
use meilisearch_types::milli::index::IndexEmbeddingConfig;
use meilisearch_types::milli::update::IndexerConfig;
use meilisearch_types::milli::vector::{Embedder, EmbedderOptions, EmbeddingConfigs};
@@ -142,6 +143,8 @@ pub struct IndexScheduler {
/// The list of tasks currently processing
pub(crate) processing_tasks: Arc<RwLock<ProcessingTasks>>,
/// The main database that also has the chat settings.
pub main: Database<Str, Unspecified>,
/// A database containing only the version of the index-scheduler
pub version: versioning::Versioning,
/// The queue containing both the tasks and the batches.
@@ -196,7 +199,7 @@ impl IndexScheduler {
version: self.version.clone(),
queue: self.queue.private_clone(),
scheduler: self.scheduler.private_clone(),
main: self.main.clone(),
index_mapper: self.index_mapper.clone(),
cleanup_enabled: self.cleanup_enabled,
webhook_url: self.webhook_url.clone(),
@@ -267,6 +270,7 @@ impl IndexScheduler {
let features = features::FeatureData::new(&env, &mut wtxn, options.instance_features)?;
let queue = Queue::new(&env, &mut wtxn, &options)?;
let index_mapper = IndexMapper::new(&env, &mut wtxn, &options, budget)?;
let chat_settings = env.create_database(&mut wtxn, Some("chat-settings"))?;
wtxn.commit()?;
// allow unreachable_code to get rids of the warning in the case of a test build.
@@ -290,6 +294,7 @@ impl IndexScheduler {
#[cfg(test)]
run_loop_iteration: Arc::new(RwLock::new(0)),
features,
chat_settings,
};
this.run();
@@ -398,9 +403,9 @@ impl IndexScheduler {
Ok(Ok(TickOutcome::StopProcessingForever)) => break,
Ok(Err(e)) => {
tracing::error!("{e}");
// Wait one second when an irrecoverable error occurs.
// Wait when an irrecoverable error occurs.
if !e.is_recoverable() {
std::thread::sleep(Duration::from_secs(1));
std::thread::sleep(Duration::from_secs(10));
}
}
Err(_panic) => {
@@ -857,6 +862,18 @@ impl IndexScheduler {
.collect();
res.map(EmbeddingConfigs::new)
}
pub fn chat_settings(&self) -> Result<Option<serde_json::Value>> {
let rtxn = self.env.read_txn().map_err(Error::HeedTransaction)?;
self.chat_settings.get(&rtxn, "main").map_err(Into::into)
}
pub fn put_chat_settings(&self, settings: &serde_json::Value) -> Result<()> {
let mut wtxn = self.env.write_txn().map_err(Error::HeedTransaction)?;
self.chat_settings.put(&mut wtxn, "main", settings)?;
wtxn.commit().map_err(Error::HeedTransaction)?;
Ok(())
}
}
/// The outcome of calling the [`IndexScheduler::tick`] function.

View File

@@ -74,6 +74,7 @@ make_enum_progress! {
make_enum_progress! {
pub enum TaskCancelationProgress {
RetrievingTasks,
CancelingUpgrade,
UpdatingTasks,
}
}

View File

@@ -423,7 +423,8 @@ impl IndexScheduler {
}
/// Create the next batch to be processed;
/// 1. We get the *last* task to cancel.
/// 0. We get the *last* task to cancel.
/// 1. We get the tasks to upgrade.
/// 2. We get the *next* task to delete.
/// 3. We get the *next* snapshot to process.
/// 4. We get the *next* dump to process.
@@ -443,7 +444,20 @@ impl IndexScheduler {
let count_total_enqueued = enqueued.len();
let failed = &self.queue.tasks.get_status(rtxn, Status::Failed)?;
// 0. The priority over everything is to upgrade the instance
// 0. we get the last task to cancel.
let to_cancel = self.queue.tasks.get_kind(rtxn, Kind::TaskCancelation)? & enqueued;
if let Some(task_id) = to_cancel.max() {
let mut task =
self.queue.tasks.get_task(rtxn, task_id)?.ok_or(Error::CorruptedTaskQueue)?;
current_batch.processing(Some(&mut task));
current_batch.reason(BatchStopReason::TaskCannotBeBatched {
kind: Kind::TaskCancelation,
id: task_id,
});
return Ok(Some((Batch::TaskCancelation { task }, current_batch)));
}
// 1. We upgrade the instance
// There shouldn't be multiple upgrade tasks but just in case we're going to batch all of them at the same time
let upgrade = self.queue.tasks.get_kind(rtxn, Kind::UpgradeDatabase)? & (enqueued | failed);
if !upgrade.is_empty() {
@@ -459,17 +473,21 @@ impl IndexScheduler {
return Ok(Some((Batch::UpgradeDatabase { tasks }, current_batch)));
}
// 1. we get the last task to cancel.
let to_cancel = self.queue.tasks.get_kind(rtxn, Kind::TaskCancelation)? & enqueued;
if let Some(task_id) = to_cancel.max() {
let mut task =
self.queue.tasks.get_task(rtxn, task_id)?.ok_or(Error::CorruptedTaskQueue)?;
current_batch.processing(Some(&mut task));
current_batch.reason(BatchStopReason::TaskCannotBeBatched {
kind: Kind::TaskCancelation,
id: task_id,
});
return Ok(Some((Batch::TaskCancelation { task }, current_batch)));
// check the version of the scheduler here.
// if the version is not the current, refuse to batch any additional task.
let version = self.version.get_version(rtxn)?;
let package_version = (
meilisearch_types::versioning::VERSION_MAJOR,
meilisearch_types::versioning::VERSION_MINOR,
meilisearch_types::versioning::VERSION_PATCH,
);
if version != Some(package_version) {
return Err(Error::UnrecoverableError(Box::new(
Error::IndexSchedulerVersionMismatch {
index_scheduler_version: version.unwrap_or((1, 12, 0)),
package_version,
},
)));
}
// 2. we get the next task to delete

View File

@@ -6,7 +6,8 @@ use meilisearch_types::batches::{BatchEnqueuedAt, BatchId};
use meilisearch_types::heed::{RoTxn, RwTxn};
use meilisearch_types::milli::progress::{Progress, VariableNameStep};
use meilisearch_types::milli::{self, ChannelCongestion};
use meilisearch_types::tasks::{Details, IndexSwap, KindWithContent, Status, Task};
use meilisearch_types::tasks::{Details, IndexSwap, Kind, KindWithContent, Status, Task};
use meilisearch_types::versioning::{VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH};
use milli::update::Settings as MilliSettings;
use roaring::RoaringBitmap;
@@ -144,11 +145,22 @@ impl IndexScheduler {
self.index_mapper.index(&rtxn, &index_uid)?
};
let mut index_wtxn = index.write_txn()?;
let index_version = index.get_version(&index_wtxn)?.unwrap_or((1, 12, 0));
let package_version = (VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH);
if index_version != package_version {
return Err(Error::IndexVersionMismatch {
index: index_uid,
index_version,
package_version,
});
}
// the index operation can take a long time, so save this handle to make it available to the search for the duration of the tick
self.index_mapper
.set_currently_updating_index(Some((index_uid.clone(), index.clone())));
let mut index_wtxn = index.write_txn()?;
let pre_commit_dabases_sizes = index.database_sizes(&index_wtxn)?;
let (tasks, congestion) =
self.apply_index_operation(&mut index_wtxn, &index, op, &progress)?;
@@ -353,9 +365,11 @@ impl IndexScheduler {
let KindWithContent::UpgradeDatabase { from } = tasks.last().unwrap().kind else {
unreachable!();
};
let ret = catch_unwind(AssertUnwindSafe(|| self.process_upgrade(from, progress)));
match ret {
Ok(Ok(())) => (),
Ok(Err(Error::AbortedTask)) => return Err(Error::AbortedTask),
Ok(Err(e)) => return Err(Error::DatabaseUpgrade(Box::new(e))),
Err(e) => {
let msg = match e.downcast_ref::<&'static str>() {
@@ -653,17 +667,79 @@ impl IndexScheduler {
progress: &Progress,
) -> Result<Vec<Task>> {
progress.update_progress(TaskCancelationProgress::RetrievingTasks);
let mut tasks_to_cancel = RoaringBitmap::new();
let enqueued_tasks = &self.queue.tasks.get_status(rtxn, Status::Enqueued)?;
// 0. Check if any upgrade task was matched.
// If so, we cancel all the failed or enqueued upgrade tasks.
let upgrade_tasks = &self.queue.tasks.get_kind(rtxn, Kind::UpgradeDatabase)?;
let is_canceling_upgrade = !matched_tasks.is_disjoint(upgrade_tasks);
if is_canceling_upgrade {
let failed_tasks = self.queue.tasks.get_status(rtxn, Status::Failed)?;
tasks_to_cancel |= upgrade_tasks & (enqueued_tasks | failed_tasks);
}
// 1. Remove from this list the tasks that we are not allowed to cancel
// Notice that only the _enqueued_ ones are cancelable and we should
// have already aborted the indexation of the _processing_ ones
let cancelable_tasks = self.queue.tasks.get_status(rtxn, Status::Enqueued)?;
let tasks_to_cancel = cancelable_tasks & matched_tasks;
tasks_to_cancel |= enqueued_tasks & matched_tasks;
// 2. If we're canceling an upgrade, attempt the rollback
if let Some(latest_upgrade_task) = (&tasks_to_cancel & upgrade_tasks).max() {
progress.update_progress(TaskCancelationProgress::CancelingUpgrade);
let task = self.queue.tasks.get_task(rtxn, latest_upgrade_task)?.unwrap();
let Some(Details::UpgradeDatabase { from, to }) = task.details else {
unreachable!("wrong details for upgrade task {latest_upgrade_task}")
};
// check that we are rollbacking an upgrade to the current Meilisearch
let bin_major: u32 = meilisearch_types::versioning::VERSION_MAJOR;
let bin_minor: u32 = meilisearch_types::versioning::VERSION_MINOR;
let bin_patch: u32 = meilisearch_types::versioning::VERSION_PATCH;
if to == (bin_major, bin_minor, bin_patch) {
tracing::warn!(
"Rollbacking from v{}.{}.{} to v{}.{}.{}",
to.0,
to.1,
to.2,
from.0,
from.1,
from.2
);
match std::panic::catch_unwind(std::panic::AssertUnwindSafe(|| {
self.process_rollback(from, progress)
})) {
Ok(Ok(())) => {}
Ok(Err(err)) => return Err(Error::DatabaseUpgrade(Box::new(err))),
Err(e) => {
let msg = match e.downcast_ref::<&'static str>() {
Some(s) => *s,
None => match e.downcast_ref::<String>() {
Some(s) => &s[..],
None => "Box<dyn Any>",
},
};
return Err(Error::DatabaseUpgrade(Box::new(Error::ProcessBatchPanicked(
msg.to_string(),
))));
}
}
} else {
tracing::debug!(
"Not rollbacking an upgrade targetting the earlier version v{}.{}.{}",
bin_major,
bin_minor,
bin_patch
)
}
}
// 3. We now have a list of tasks to cancel, cancel them
let (task_progress, progress_obj) = AtomicTaskStep::new(tasks_to_cancel.len() as u32);
progress.update_progress(progress_obj);
// 2. We now have a list of tasks to cancel, cancel them
let mut tasks = self.queue.tasks.get_existing_tasks(
rtxn,
tasks_to_cancel.iter().inspect(|_| {

View File

@@ -12,10 +12,14 @@ impl IndexScheduler {
#[cfg(test)]
self.maybe_fail(crate::test_utils::FailureLocation::ProcessUpgrade)?;
enum UpgradeIndex {}
let indexes = self.index_names()?;
for (i, uid) in indexes.iter().enumerate() {
let must_stop_processing = self.scheduler.must_stop_processing.clone();
if must_stop_processing.get() {
return Err(Error::AbortedTask);
}
progress.update_progress(VariableNameStep::<UpgradeIndex>::new(
format!("Upgrading index `{uid}`"),
i as u32,
@@ -27,6 +31,7 @@ impl IndexScheduler {
&mut index_wtxn,
&index,
db_version,
|| must_stop_processing.get(),
progress.clone(),
)
.map_err(|e| Error::from_milli(e, Some(uid.to_string())))?;
@@ -46,4 +51,42 @@ impl IndexScheduler {
Ok(())
}
pub fn process_rollback(&self, db_version: (u32, u32, u32), progress: &Progress) -> Result<()> {
let mut wtxn = self.env.write_txn()?;
tracing::info!(?db_version, "roll back index scheduler version");
self.version.set_version(&mut wtxn, db_version)?;
let db_path = self.scheduler.version_file_path.parent().unwrap();
wtxn.commit()?;
let indexes = self.index_names()?;
tracing::info!("roll backing all indexes");
for (i, uid) in indexes.iter().enumerate() {
progress.update_progress(VariableNameStep::<UpgradeIndex>::new(
format!("Rollbacking index `{uid}`"),
i as u32,
indexes.len() as u32,
));
let index_schd_rtxn = self.env.read_txn()?;
let rollback_outcome =
self.index_mapper.rollback_index(&index_schd_rtxn, uid, db_version)?;
if !rollback_outcome.succeeded() {
return Err(crate::Error::RollbackFailed { index: uid.clone(), rollback_outcome });
}
}
tracing::info!(?db_path, ?db_version, "roll back version file");
meilisearch_types::versioning::create_version_file(
db_path,
db_version.0,
db_version.1,
db_version.2,
)?;
Ok(())
}
}
enum UpgradeIndex {}

View File

@@ -6,7 +6,7 @@ source: crates/index-scheduler/src/scheduler/test_failure.rs
[]
----------------------------------------------------------------------
### All Tasks:
0 {uid: 0, batch_uid: 0, status: succeeded, details: { from: (1, 12, 0), to: (1, 14, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
0 {uid: 0, batch_uid: 0, status: succeeded, details: { from: (1, 12, 0), to: (1, 15, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
1 {uid: 1, batch_uid: 1, status: succeeded, details: { primary_key: Some("mouse") }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}
2 {uid: 2, batch_uid: 2, status: succeeded, details: { primary_key: Some("bone") }, kind: IndexCreation { index_uid: "doggo", primary_key: Some("bone") }}
3 {uid: 3, batch_uid: 3, status: failed, error: ResponseError { code: 200, message: "Index `doggo` already exists.", error_code: "index_already_exists", error_type: "invalid_request", error_link: "https://docs.meilisearch.com/errors#index_already_exists" }, details: { primary_key: Some("bone") }, kind: IndexCreation { index_uid: "doggo", primary_key: Some("bone") }}
@@ -57,7 +57,7 @@ girafo: { number_of_documents: 0, field_distribution: {} }
[timestamp] [4,]
----------------------------------------------------------------------
### All Batches:
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.14.0"}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task", }
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.15.0"}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task", }
1 {uid: 1, details: {"primaryKey":"mouse"}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"indexCreation":1},"indexUids":{"catto":1}}, stop reason: "task with id 1 of type `indexCreation` cannot be batched", }
2 {uid: 2, details: {"primaryKey":"bone"}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"indexCreation":1},"indexUids":{"doggo":1}}, stop reason: "task with id 2 of type `indexCreation` cannot be batched", }
3 {uid: 3, details: {"primaryKey":"bone"}, stats: {"totalNbTasks":1,"status":{"failed":1},"types":{"indexCreation":1},"indexUids":{"doggo":1}}, stop reason: "task with id 3 of type `indexCreation` cannot be batched", }

View File

@@ -6,7 +6,7 @@ source: crates/index-scheduler/src/scheduler/test_failure.rs
[]
----------------------------------------------------------------------
### All Tasks:
0 {uid: 0, status: enqueued, details: { from: (1, 12, 0), to: (1, 14, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
0 {uid: 0, status: enqueued, details: { from: (1, 12, 0), to: (1, 15, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
----------------------------------------------------------------------
### Status:
enqueued [0,]

View File

@@ -6,7 +6,7 @@ source: crates/index-scheduler/src/scheduler/test_failure.rs
[]
----------------------------------------------------------------------
### All Tasks:
0 {uid: 0, status: enqueued, details: { from: (1, 12, 0), to: (1, 14, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
0 {uid: 0, status: enqueued, details: { from: (1, 12, 0), to: (1, 15, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
1 {uid: 1, status: enqueued, details: { primary_key: Some("mouse") }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}
----------------------------------------------------------------------
### Status:

View File

@@ -6,7 +6,7 @@ source: crates/index-scheduler/src/scheduler/test_failure.rs
[]
----------------------------------------------------------------------
### All Tasks:
0 {uid: 0, batch_uid: 0, status: failed, error: ResponseError { code: 200, message: "Planned failure for tests.", error_code: "internal", error_type: "internal", error_link: "https://docs.meilisearch.com/errors#internal" }, details: { from: (1, 12, 0), to: (1, 14, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
0 {uid: 0, batch_uid: 0, status: failed, error: ResponseError { code: 200, message: "Planned failure for tests.", error_code: "internal", error_type: "internal", error_link: "https://docs.meilisearch.com/errors#internal" }, details: { from: (1, 12, 0), to: (1, 15, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
1 {uid: 1, status: enqueued, details: { primary_key: Some("mouse") }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}
----------------------------------------------------------------------
### Status:
@@ -37,7 +37,7 @@ catto [1,]
[timestamp] [0,]
----------------------------------------------------------------------
### All Batches:
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.14.0"}, stats: {"totalNbTasks":1,"status":{"failed":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task", }
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.15.0"}, stats: {"totalNbTasks":1,"status":{"failed":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task", }
----------------------------------------------------------------------
### Batch to tasks mapping:
0 [0,]

View File

@@ -6,7 +6,7 @@ source: crates/index-scheduler/src/scheduler/test_failure.rs
[]
----------------------------------------------------------------------
### All Tasks:
0 {uid: 0, batch_uid: 0, status: failed, error: ResponseError { code: 200, message: "Planned failure for tests.", error_code: "internal", error_type: "internal", error_link: "https://docs.meilisearch.com/errors#internal" }, details: { from: (1, 12, 0), to: (1, 14, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
0 {uid: 0, batch_uid: 0, status: failed, error: ResponseError { code: 200, message: "Planned failure for tests.", error_code: "internal", error_type: "internal", error_link: "https://docs.meilisearch.com/errors#internal" }, details: { from: (1, 12, 0), to: (1, 15, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
1 {uid: 1, status: enqueued, details: { primary_key: Some("mouse") }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}
2 {uid: 2, status: enqueued, details: { primary_key: Some("bone") }, kind: IndexCreation { index_uid: "doggo", primary_key: Some("bone") }}
----------------------------------------------------------------------
@@ -40,7 +40,7 @@ doggo [2,]
[timestamp] [0,]
----------------------------------------------------------------------
### All Batches:
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.14.0"}, stats: {"totalNbTasks":1,"status":{"failed":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task", }
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.15.0"}, stats: {"totalNbTasks":1,"status":{"failed":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task", }
----------------------------------------------------------------------
### Batch to tasks mapping:
0 [0,]

View File

@@ -6,7 +6,7 @@ source: crates/index-scheduler/src/scheduler/test_failure.rs
[]
----------------------------------------------------------------------
### All Tasks:
0 {uid: 0, batch_uid: 0, status: succeeded, details: { from: (1, 12, 0), to: (1, 14, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
0 {uid: 0, batch_uid: 0, status: succeeded, details: { from: (1, 12, 0), to: (1, 15, 0) }, kind: UpgradeDatabase { from: (1, 12, 0) }}
1 {uid: 1, status: enqueued, details: { primary_key: Some("mouse") }, kind: IndexCreation { index_uid: "catto", primary_key: Some("mouse") }}
2 {uid: 2, status: enqueued, details: { primary_key: Some("bone") }, kind: IndexCreation { index_uid: "doggo", primary_key: Some("bone") }}
3 {uid: 3, status: enqueued, details: { primary_key: Some("bone") }, kind: IndexCreation { index_uid: "doggo", primary_key: Some("bone") }}
@@ -43,7 +43,7 @@ doggo [2,3,]
[timestamp] [0,]
----------------------------------------------------------------------
### All Batches:
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.14.0"}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task", }
0 {uid: 0, details: {"upgradeFrom":"v1.12.0","upgradeTo":"v1.15.0"}, stats: {"totalNbTasks":1,"status":{"succeeded":1},"types":{"upgradeDatabase":1},"indexUids":{}}, stop reason: "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task", }
----------------------------------------------------------------------
### Batch to tasks mapping:
0 [0,]

View File

@@ -0,0 +1,432 @@
use std::collections::{BTreeMap, BTreeSet};
use std::convert::Infallible;
use std::fmt;
use std::marker::PhantomData;
use std::num::NonZeroUsize;
use std::ops::{ControlFlow, Deref};
use std::str::FromStr;
use deserr::{DeserializeError, Deserr, ErrorKind, MergeWithError, ValuePointerRef};
use fst::IntoStreamer;
use milli::disabled_typos_terms::DisabledTyposTerms;
use milli::index::{IndexEmbeddingConfig, PrefixSearch};
use milli::proximity::ProximityPrecision;
use milli::update::Setting;
use milli::{FilterableAttributesRule, Index};
use serde::{Deserialize, Serialize, Serializer};
use utoipa::ToSchema;
use crate::deserr::DeserrJsonError;
use crate::error::deserr_codes::*;
use crate::heed::RoTxn;
use crate::IndexScheduler;
#[derive(Debug, Clone, Default, Serialize, Deserialize, PartialEq, Eq, Deserr, ToSchema)]
#[serde(deny_unknown_fields, rename_all = "camelCase")]
#[deserr(deny_unknown_fields, rename_all = camelCase, where_predicate = __Deserr_E: deserr::MergeWithError<DeserrJsonError<InvalidSettingsTypoTolerance>>)]
pub struct PromptsSettings {
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<String>)]
pub system: Setting<String>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsTypoTolerance>)]
#[schema(value_type = Option<MinWordSizeTyposSetting>, example = json!({ "oneTypo": 5, "twoTypo": 9 }))]
pub search_description: Setting<String>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<String>)]
pub search_q_param: Setting<String>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<String>)]
pub pre_query: Setting<String>,
}
#[derive(Debug, Clone, Default, Serialize, Deserialize, PartialEq, Eq, Deserr, ToSchema)]
#[serde(deny_unknown_fields, rename_all = "camelCase")]
pub enum ChatSource {
#[default]
OpenAi,
}
/// Holds all the settings for an index. `T` can either be `Checked` if they represents settings
/// whose validity is guaranteed, or `Unchecked` if they need to be validated. In the later case, a
/// call to `check` will return a `Settings<Checked>` from a `Settings<Unchecked>`.
#[derive(Debug, Clone, Default, Serialize, Deserialize, PartialEq, Eq, Deserr, ToSchema)]
#[serde(
deny_unknown_fields,
rename_all = "camelCase",
bound(serialize = "T: Serialize", deserialize = "T: Deserialize<'static>")
)]
#[deserr(error = DeserrJsonError, rename_all = camelCase, deny_unknown_fields)]
#[schema(rename_all = "camelCase")]
pub struct ChatSettings<T> {
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsDisplayedAttributes>)]
#[schema(value_type = Option<String>)]
pub source: Setting<ChatSource>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsSearchableAttributes>)]
#[schema(value_type = Option<String>)]
pub base_api: Setting<String>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsSearchableAttributes>)]
#[schema(value_type = Option<String>)]
pub api_key: Setting<String>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsFilterableAttributes>)]
#[schema(value_type = Option<PromptsSettings>)]
pub prompts: Setting<PromptsSettings>,
#[serde(skip)]
#[deserr(skip)]
pub _kind: PhantomData<T>,
}
impl<T> ChatSettings<T> {
pub fn hide_secrets(&mut self) {
match &mut self.api_key {
Setting::Set(key) => Self::hide_secrets(key),
Setting::Reset => todo!(),
Setting::NotSet => todo!(),
}
}
fn hide_secret(secret: &mut String) {
match secret.len() {
x if x < 10 => {
secret.replace_range(.., "XXX...");
}
x if x < 20 => {
secret.replace_range(2.., "XXXX...");
}
x if x < 30 => {
secret.replace_range(3.., "XXXXX...");
}
_x => {
secret.replace_range(5.., "XXXXXX...");
}
}
}
}
impl ChatSettings<Checked> {
pub fn cleared() -> ChatSettings<Checked> {
ChatSettings {
source: Setting::Reset,
base_api: Setting::Reset,
api_key: Setting::Reset,
prompts: Setting::Reset,
_kind: PhantomData,
}
}
pub fn into_unchecked(self) -> ChatSettings<Unchecked> {
let Self { source, base_api, api_key, prompts, _kind } = self;
ChatSettings { source, base_api, api_key, prompts, _kind: PhantomData }
}
}
impl ChatSettings<Unchecked> {
pub fn check(self) -> ChatSettings<Checked> {
ChatSettings {
source: self.source,
base_api: self.base_api,
api_key: self.api_key,
prompts: self.prompts,
_kind: PhantomData,
}
}
pub fn validate(self) -> Result<Self, milli::Error> {
self.validate_prompt_settings()?;
self.validate_global_settings()
}
fn validate_global_settings(mut self) -> Result<Self, milli::Error> {
// Check that the ApiBase is a valid URL
Ok(self)
}
fn validate_prompt_settings(mut self) -> Result<Self, milli::Error> {
// TODO
// let Setting::Set(mut configs) = self.embedders else { return Ok(self) };
// for (name, config) in configs.iter_mut() {
// let config_to_check = std::mem::take(config);
// let checked_config =
// milli::update::validate_embedding_settings(config_to_check.inner, name)?;
// *config = SettingEmbeddingSettings { inner: checked_config };
// }
// self.embedders = Setting::Set(configs);
Ok(self)
}
pub fn merge(&mut self, other: &Self) {
// For most settings only the latest version is kept
*self = Self {
source: other.source.or(self.source),
base_api: other.base_api.or(self.base_api),
api_key: other.api_key.or(self.api_key),
prompts: match (self.prompts, other.prompts) {
(Setting::NotSet, set) | (set, Setting::NotSet) => set,
(Setting::Set(_) | Setting::Reset, Setting::Reset) => Setting::Reset,
(Setting::Reset, Setting::Set(set)) => Setting::Set(set),
// If both are set we must merge the prompts settings
(Setting::Set(this), Setting::Set(other)) => Setting::Set(PromptsSettings {
system: other.system.or(system),
search_description: other.search_description.or(search_description),
search_q_param: other.search_q_param.or(search_q_param),
pre_query: other.pre_query.or(pre_query),
}),
},
_kind: PhantomData,
}
}
}
pub fn apply_settings_to_builder(
settings: &ChatSettings<Checked>,
// TODO we must not store this into milli but in the index scheduler
builder: &mut milli::update::Settings,
) {
let ChatSettings { source, base_api, api_key, prompts, _kind } = settings;
match source.deref() {
Setting::Set(ref names) => builder.set_searchable_fields(names.clone()),
Setting::Reset => builder.reset_searchable_fields(),
Setting::NotSet => (),
}
match displayed_attributes.deref() {
Setting::Set(ref names) => builder.set_displayed_fields(names.clone()),
Setting::Reset => builder.reset_displayed_fields(),
Setting::NotSet => (),
}
match filterable_attributes {
Setting::Set(ref facets) => {
builder.set_filterable_fields(facets.clone().into_iter().collect())
}
Setting::Reset => builder.reset_filterable_fields(),
Setting::NotSet => (),
}
match sortable_attributes {
Setting::Set(ref fields) => builder.set_sortable_fields(fields.iter().cloned().collect()),
Setting::Reset => builder.reset_sortable_fields(),
Setting::NotSet => (),
}
match ranking_rules {
Setting::Set(ref criteria) => {
builder.set_criteria(criteria.iter().map(|c| c.clone().into()).collect())
}
Setting::Reset => builder.reset_criteria(),
Setting::NotSet => (),
}
match stop_words {
Setting::Set(ref stop_words) => builder.set_stop_words(stop_words.clone()),
Setting::Reset => builder.reset_stop_words(),
Setting::NotSet => (),
}
match non_separator_tokens {
Setting::Set(ref non_separator_tokens) => {
builder.set_non_separator_tokens(non_separator_tokens.clone())
}
Setting::Reset => builder.reset_non_separator_tokens(),
Setting::NotSet => (),
}
match separator_tokens {
Setting::Set(ref separator_tokens) => {
builder.set_separator_tokens(separator_tokens.clone())
}
Setting::Reset => builder.reset_separator_tokens(),
Setting::NotSet => (),
}
match dictionary {
Setting::Set(ref dictionary) => builder.set_dictionary(dictionary.clone()),
Setting::Reset => builder.reset_dictionary(),
Setting::NotSet => (),
}
match synonyms {
Setting::Set(ref synonyms) => builder.set_synonyms(synonyms.clone().into_iter().collect()),
Setting::Reset => builder.reset_synonyms(),
Setting::NotSet => (),
}
match distinct_attribute {
Setting::Set(ref attr) => builder.set_distinct_field(attr.clone()),
Setting::Reset => builder.reset_distinct_field(),
Setting::NotSet => (),
}
match proximity_precision {
Setting::Set(ref precision) => builder.set_proximity_precision((*precision).into()),
Setting::Reset => builder.reset_proximity_precision(),
Setting::NotSet => (),
}
match localized_attributes_rules {
Setting::Set(ref rules) => builder
.set_localized_attributes_rules(rules.iter().cloned().map(|r| r.into()).collect()),
Setting::Reset => builder.reset_localized_attributes_rules(),
Setting::NotSet => (),
}
match typo_tolerance {
Setting::Set(ref value) => {
match value.enabled {
Setting::Set(val) => builder.set_autorize_typos(val),
Setting::Reset => builder.reset_authorize_typos(),
Setting::NotSet => (),
}
match value.min_word_size_for_typos {
Setting::Set(ref setting) => {
match setting.one_typo {
Setting::Set(val) => builder.set_min_word_len_one_typo(val),
Setting::Reset => builder.reset_min_word_len_one_typo(),
Setting::NotSet => (),
}
match setting.two_typos {
Setting::Set(val) => builder.set_min_word_len_two_typos(val),
Setting::Reset => builder.reset_min_word_len_two_typos(),
Setting::NotSet => (),
}
}
Setting::Reset => {
builder.reset_min_word_len_one_typo();
builder.reset_min_word_len_two_typos();
}
Setting::NotSet => (),
}
match value.disable_on_words {
Setting::Set(ref words) => {
builder.set_exact_words(words.clone());
}
Setting::Reset => builder.reset_exact_words(),
Setting::NotSet => (),
}
match value.disable_on_attributes {
Setting::Set(ref words) => {
builder.set_exact_attributes(words.iter().cloned().collect())
}
Setting::Reset => builder.reset_exact_attributes(),
Setting::NotSet => (),
}
match value.disable_on_numbers {
Setting::Set(val) => builder.set_disable_on_numbers(val),
Setting::Reset => builder.reset_disable_on_numbers(),
Setting::NotSet => (),
}
}
Setting::Reset => {
// all typo settings need to be reset here.
builder.reset_authorize_typos();
builder.reset_min_word_len_one_typo();
builder.reset_min_word_len_two_typos();
builder.reset_exact_words();
builder.reset_exact_attributes();
}
Setting::NotSet => (),
}
match faceting {
Setting::Set(FacetingSettings { max_values_per_facet, sort_facet_values_by }) => {
match max_values_per_facet {
Setting::Set(val) => builder.set_max_values_per_facet(*val),
Setting::Reset => builder.reset_max_values_per_facet(),
Setting::NotSet => (),
}
match sort_facet_values_by {
Setting::Set(val) => builder.set_sort_facet_values_by(
val.iter().map(|(name, order)| (name.clone(), (*order).into())).collect(),
),
Setting::Reset => builder.reset_sort_facet_values_by(),
Setting::NotSet => (),
}
}
Setting::Reset => {
builder.reset_max_values_per_facet();
builder.reset_sort_facet_values_by();
}
Setting::NotSet => (),
}
match pagination {
Setting::Set(ref value) => match value.max_total_hits {
Setting::Set(val) => builder.set_pagination_max_total_hits(val),
Setting::Reset => builder.reset_pagination_max_total_hits(),
Setting::NotSet => (),
},
Setting::Reset => builder.reset_pagination_max_total_hits(),
Setting::NotSet => (),
}
match embedders {
Setting::Set(value) => builder.set_embedder_settings(
value.iter().map(|(k, v)| (k.clone(), v.inner.clone())).collect(),
),
Setting::Reset => builder.reset_embedder_settings(),
Setting::NotSet => (),
}
match search_cutoff_ms {
Setting::Set(cutoff) => builder.set_search_cutoff(*cutoff),
Setting::Reset => builder.reset_search_cutoff(),
Setting::NotSet => (),
}
match prefix_search {
Setting::Set(prefix_search) => {
builder.set_prefix_search(PrefixSearch::from(*prefix_search))
}
Setting::Reset => builder.reset_prefix_search(),
Setting::NotSet => (),
}
match facet_search {
Setting::Set(facet_search) => builder.set_facet_search(*facet_search),
Setting::Reset => builder.reset_facet_search(),
Setting::NotSet => (),
}
match chat {
Setting::Set(chat) => builder.set_chat(chat.clone()),
Setting::Reset => builder.reset_chat(),
Setting::NotSet => (),
}
}
pub enum SecretPolicy {
RevealSecrets,
HideSecrets,
}
pub fn settings(
index_scheduler: &IndexScheduler,
rtxn: &RoTxn,
secret_policy: SecretPolicy,
) -> Result<Settings<Checked>, milli::Error> {
let mut settings = index_scheduler.chat_settings(rtxn)?;
if let SecretPolicy::HideSecrets = secret_policy {
settings.hide_secrets()
}
Ok(settings)
}

View File

@@ -0,0 +1,3 @@
mod chat;
pub use chat::ChatSettings;

View File

@@ -114,12 +114,8 @@ impl IndexScheduler {
auto_upgrade: true, // Don't cost much and will ensure the happy path works
embedding_cache_cap: 10,
};
let version = configuration(&mut options).unwrap_or_else(|| {
(
versioning::VERSION_MAJOR.parse().unwrap(),
versioning::VERSION_MINOR.parse().unwrap(),
versioning::VERSION_PATCH.parse().unwrap(),
)
let version = configuration(&mut options).unwrap_or({
(versioning::VERSION_MAJOR, versioning::VERSION_MINOR, versioning::VERSION_PATCH)
});
std::fs::create_dir_all(&options.auth_path).unwrap();

View File

@@ -28,12 +28,17 @@ pub fn upgrade_index_scheduler(
let current_minor = to.1;
let current_patch = to.2;
let upgrade_functions: &[&dyn UpgradeIndexScheduler] = &[&ToCurrentNoOp {}];
let upgrade_functions: &[&dyn UpgradeIndexScheduler] = &[
// This is the last upgrade function, it will be called when the index is up to date.
// any other upgrade function should be added before this one.
&ToCurrentNoOp {},
];
let start = match from {
(1, 12, _) => 0,
(1, 13, _) => 0,
(1, 14, _) => 0,
(1, 15, _) => 0,
(major, minor, patch) => {
if major > current_major
|| (major == current_major && minor > current_minor)
@@ -104,10 +109,6 @@ impl UpgradeIndexScheduler for ToCurrentNoOp {
}
fn target_version(&self) -> (u32, u32, u32) {
(
VERSION_MAJOR.parse().unwrap(),
VERSION_MINOR.parse().unwrap(),
VERSION_PATCH.parse().unwrap(),
)
(VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH)
}
}

View File

@@ -39,9 +39,9 @@ impl Versioning {
}
pub fn set_current_version(&self, wtxn: &mut RwTxn) -> Result<(), heed::Error> {
let major = versioning::VERSION_MAJOR.parse().unwrap();
let minor = versioning::VERSION_MINOR.parse().unwrap();
let patch = versioning::VERSION_PATCH.parse().unwrap();
let major = versioning::VERSION_MAJOR;
let minor = versioning::VERSION_MINOR;
let patch = versioning::VERSION_PATCH;
self.set_version(wtxn, (major, minor, patch))
}
@@ -64,9 +64,9 @@ impl Versioning {
};
wtxn.commit()?;
let bin_major: u32 = versioning::VERSION_MAJOR.parse().unwrap();
let bin_minor: u32 = versioning::VERSION_MINOR.parse().unwrap();
let bin_patch: u32 = versioning::VERSION_PATCH.parse().unwrap();
let bin_major: u32 = versioning::VERSION_MAJOR;
let bin_minor: u32 = versioning::VERSION_MINOR;
let bin_patch: u32 = versioning::VERSION_PATCH;
let to = (bin_major, bin_minor, bin_patch);
if from != to {

View File

@@ -351,6 +351,7 @@ pub struct IndexSearchRules {
fn generate_default_keys(store: &HeedAuthStore) -> Result<()> {
store.put_api_key(Key::default_admin())?;
store.put_api_key(Key::default_search())?;
store.put_api_key(Key::default_chat())?;
Ok(())
}

View File

@@ -22,7 +22,7 @@ pub struct BatchView {
#[serde(with = "time::serde::rfc3339::option", default)]
pub finished_at: Option<OffsetDateTime>,
#[serde(default = "meilisearch_types::batches::default_stop_reason")]
pub batcher_stopped_because: String,
pub batch_creation_complete: String,
}
impl BatchView {
@@ -35,7 +35,7 @@ impl BatchView {
duration: batch.finished_at.map(|finished_at| finished_at - batch.started_at),
started_at: batch.started_at,
finished_at: batch.finished_at,
batcher_stopped_because: batch.stop_reason.clone(),
batch_creation_complete: batch.stop_reason.clone(),
}
}
}

View File

@@ -387,7 +387,8 @@ VectorEmbeddingError , InvalidRequest , BAD_REQUEST ;
NotFoundSimilarId , InvalidRequest , BAD_REQUEST ;
InvalidDocumentEditionContext , InvalidRequest , BAD_REQUEST ;
InvalidDocumentEditionFunctionFilter , InvalidRequest , BAD_REQUEST ;
EditDocumentsByFunctionError , InvalidRequest , BAD_REQUEST
EditDocumentsByFunctionError , InvalidRequest , BAD_REQUEST ;
InvalidSettingsIndexChat , InvalidRequest , BAD_REQUEST
}
impl ErrorCode for JoinError {

View File

@@ -158,6 +158,21 @@ impl Key {
updated_at: now,
}
}
pub fn default_chat() -> Self {
let now = OffsetDateTime::now_utc();
let uid = Uuid::new_v4();
Self {
name: Some("Default Chat API Key".to_string()),
description: Some("Use it to chat and search from the frontend".to_string()),
uid,
actions: vec![Action::Chat, Action::Search],
indexes: vec![IndexUidPattern::all()],
expires_at: None,
created_at: now,
updated_at: now,
}
}
}
fn parse_expiration_date(
@@ -308,6 +323,18 @@ pub enum Action {
#[serde(rename = "network.update")]
#[deserr(rename = "network.update")]
NetworkUpdate,
#[serde(rename = "chat.get")]
#[deserr(rename = "chat.get")]
Chat,
#[serde(rename = "chatSettings.*")]
#[deserr(rename = "chatSettings.*")]
ChatSettingsAll,
#[serde(rename = "chatSettings.get")]
#[deserr(rename = "chatSettings.get")]
ChatSettingsGet,
#[serde(rename = "chatSettings.update")]
#[deserr(rename = "chatSettings.update")]
ChatSettingsUpdate,
}
impl Action {
@@ -333,6 +360,9 @@ impl Action {
SETTINGS_ALL => Some(Self::SettingsAll),
SETTINGS_GET => Some(Self::SettingsGet),
SETTINGS_UPDATE => Some(Self::SettingsUpdate),
CHAT_SETTINGS_ALL => Some(Self::ChatSettingsAll),
CHAT_SETTINGS_GET => Some(Self::ChatSettingsGet),
CHAT_SETTINGS_UPDATE => Some(Self::ChatSettingsUpdate),
STATS_ALL => Some(Self::StatsAll),
STATS_GET => Some(Self::StatsGet),
METRICS_ALL => Some(Self::MetricsAll),
@@ -349,6 +379,7 @@ impl Action {
EXPERIMENTAL_FEATURES_UPDATE => Some(Self::ExperimentalFeaturesUpdate),
NETWORK_GET => Some(Self::NetworkGet),
NETWORK_UPDATE => Some(Self::NetworkUpdate),
CHAT => Some(Self::Chat),
_otherwise => None,
}
}
@@ -397,4 +428,9 @@ pub mod actions {
pub const NETWORK_GET: u8 = NetworkGet.repr();
pub const NETWORK_UPDATE: u8 = NetworkUpdate.repr();
pub const CHAT: u8 = Chat.repr();
pub const CHAT_SETTINGS_ALL: u8 = ChatSettingsAll.repr();
pub const CHAT_SETTINGS_GET: u8 = ChatSettingsGet.repr();
pub const CHAT_SETTINGS_UPDATE: u8 = ChatSettingsUpdate.repr();
}

View File

@@ -8,13 +8,16 @@ use std::str::FromStr;
use deserr::{DeserializeError, Deserr, ErrorKind, MergeWithError, ValuePointerRef};
use fst::IntoStreamer;
use milli::disabled_typos_terms::DisabledTyposTerms;
use milli::index::{IndexEmbeddingConfig, PrefixSearch};
use milli::proximity::ProximityPrecision;
pub use milli::update::ChatSettings;
use milli::update::Setting;
use milli::{Criterion, CriterionError, FilterableAttributesRule, Index, DEFAULT_VALUES_PER_FACET};
use serde::{Deserialize, Serialize, Serializer};
use utoipa::ToSchema;
use super::{Checked, Unchecked};
use crate::deserr::DeserrJsonError;
use crate::error::deserr_codes::*;
use crate::facet_values_sort::FacetValuesSort;
@@ -104,6 +107,10 @@ pub struct TypoSettings {
#[deserr(default)]
#[schema(value_type = Option<BTreeSet<String>>, example = json!(["uuid", "url"]))]
pub disable_on_attributes: Setting<BTreeSet<String>>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<bool>, example = json!(true))]
pub disable_on_numbers: Setting<bool>,
}
#[derive(Debug, Clone, Default, Serialize, Deserialize, PartialEq, Eq, Deserr, ToSchema)]
@@ -194,72 +201,86 @@ pub struct Settings<T> {
#[deserr(default, error = DeserrJsonError<InvalidSettingsDisplayedAttributes>)]
#[schema(value_type = Option<Vec<String>>, example = json!(["id", "title", "description", "url"]))]
pub displayed_attributes: WildcardSetting,
/// Fields in which to search for matching query words sorted by order of importance.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsSearchableAttributes>)]
#[schema(value_type = Option<Vec<String>>, example = json!(["title", "description"]))]
pub searchable_attributes: WildcardSetting,
/// Attributes to use for faceting and filtering. See [Filtering and Faceted Search](https://www.meilisearch.com/docs/learn/filtering_and_sorting/search_with_facet_filters).
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsFilterableAttributes>)]
#[schema(value_type = Option<Vec<FilterableAttributesRule>>, example = json!(["release_date", "genre"]))]
pub filterable_attributes: Setting<Vec<FilterableAttributesRule>>,
/// Attributes to use when sorting search results.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsSortableAttributes>)]
#[schema(value_type = Option<Vec<String>>, example = json!(["release_date"]))]
pub sortable_attributes: Setting<BTreeSet<String>>,
/// List of ranking rules sorted by order of importance. The order is customizable.
/// [A list of ordered built-in ranking rules](https://www.meilisearch.com/docs/learn/relevancy/relevancy).
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsRankingRules>)]
#[schema(value_type = Option<Vec<String>>, example = json!([RankingRuleView::Words, RankingRuleView::Typo, RankingRuleView::Proximity, RankingRuleView::Attribute, RankingRuleView::Exactness]))]
pub ranking_rules: Setting<Vec<RankingRuleView>>,
/// List of words ignored when present in search queries.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsStopWords>)]
#[schema(value_type = Option<Vec<String>>, example = json!(["the", "a", "them", "their"]))]
pub stop_words: Setting<BTreeSet<String>>,
/// List of characters not delimiting where one term begins and ends.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsNonSeparatorTokens>)]
#[schema(value_type = Option<Vec<String>>, example = json!([" ", "\n"]))]
pub non_separator_tokens: Setting<BTreeSet<String>>,
/// List of characters delimiting where one term begins and ends.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsSeparatorTokens>)]
#[schema(value_type = Option<Vec<String>>, example = json!(["S"]))]
pub separator_tokens: Setting<BTreeSet<String>>,
/// List of strings Meilisearch should parse as a single term.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsDictionary>)]
#[schema(value_type = Option<Vec<String>>, example = json!(["iPhone pro"]))]
pub dictionary: Setting<BTreeSet<String>>,
/// List of associated words treated similarly. A word associated to an array of word as synonyms.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsSynonyms>)]
#[schema(value_type = Option<BTreeMap<String, Vec<String>>>, example = json!({ "he": ["she", "they", "them"], "phone": ["iPhone", "android"]}))]
pub synonyms: Setting<BTreeMap<String, Vec<String>>>,
/// Search returns documents with distinct (different) values of the given field.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsDistinctAttribute>)]
#[schema(value_type = Option<String>, example = json!("sku"))]
pub distinct_attribute: Setting<String>,
/// Precision level when calculating the proximity ranking rule.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsProximityPrecision>)]
#[schema(value_type = Option<String>, example = json!(ProximityPrecisionView::ByAttribute))]
pub proximity_precision: Setting<ProximityPrecisionView>,
/// Customize typo tolerance feature.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsTypoTolerance>)]
#[schema(value_type = Option<TypoSettings>, example = json!({ "enabled": true, "disableOnAttributes": ["title"]}))]
pub typo_tolerance: Setting<TypoSettings>,
/// Faceting settings.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsFaceting>)]
#[schema(value_type = Option<FacetingSettings>, example = json!({ "maxValuesPerFacet": 10, "sortFacetValuesBy": { "genre": FacetValuesSort::Count }}))]
pub faceting: Setting<FacetingSettings>,
/// Pagination settings.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsPagination>)]
@@ -271,24 +292,34 @@ pub struct Settings<T> {
#[deserr(default, error = DeserrJsonError<InvalidSettingsEmbedders>)]
#[schema(value_type = Option<BTreeMap<String, SettingEmbeddingSettings>>)]
pub embedders: Setting<BTreeMap<String, SettingEmbeddingSettings>>,
/// Maximum duration of a search query.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsSearchCutoffMs>)]
#[schema(value_type = Option<u64>, example = json!(50))]
pub search_cutoff_ms: Setting<u64>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsLocalizedAttributes>)]
#[schema(value_type = Option<Vec<LocalizedAttributesRuleView>>, example = json!(50))]
pub localized_attributes: Setting<Vec<LocalizedAttributesRuleView>>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsFacetSearch>)]
#[schema(value_type = Option<bool>, example = json!(true))]
pub facet_search: Setting<bool>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsPrefixSearch>)]
#[schema(value_type = Option<PrefixSearchSettings>, example = json!("Hemlo"))]
pub prefix_search: Setting<PrefixSearchSettings>,
/// Customize the chat prompting.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default, error = DeserrJsonError<InvalidSettingsIndexChat>)]
#[schema(value_type = Option<ChatSettings>)]
pub chat: Setting<ChatSettings>,
#[serde(skip)]
#[deserr(skip)]
pub _kind: PhantomData<T>,
@@ -354,6 +385,7 @@ impl Settings<Checked> {
localized_attributes: Setting::Reset,
facet_search: Setting::Reset,
prefix_search: Setting::Reset,
chat: Setting::Reset,
_kind: PhantomData,
}
}
@@ -380,6 +412,7 @@ impl Settings<Checked> {
localized_attributes: localized_attributes_rules,
facet_search,
prefix_search,
chat,
_kind,
} = self;
@@ -404,6 +437,7 @@ impl Settings<Checked> {
localized_attributes: localized_attributes_rules,
facet_search,
prefix_search,
chat,
_kind: PhantomData,
}
}
@@ -454,6 +488,7 @@ impl Settings<Unchecked> {
localized_attributes: self.localized_attributes,
facet_search: self.facet_search,
prefix_search: self.prefix_search,
chat: self.chat,
_kind: PhantomData,
}
}
@@ -528,8 +563,9 @@ impl Settings<Unchecked> {
Setting::Set(this)
}
},
prefix_search: other.prefix_search.or(self.prefix_search),
facet_search: other.facet_search.or(self.facet_search),
prefix_search: other.prefix_search.or(self.prefix_search),
chat: other.chat.clone().or(self.chat.clone()),
_kind: PhantomData,
}
}
@@ -568,6 +604,7 @@ pub fn apply_settings_to_builder(
localized_attributes: localized_attributes_rules,
facet_search,
prefix_search,
chat,
_kind,
} = settings;
@@ -701,6 +738,12 @@ pub fn apply_settings_to_builder(
Setting::Reset => builder.reset_exact_attributes(),
Setting::NotSet => (),
}
match value.disable_on_numbers {
Setting::Set(val) => builder.set_disable_on_numbers(val),
Setting::Reset => builder.reset_disable_on_numbers(),
Setting::NotSet => (),
}
}
Setting::Reset => {
// all typo settings need to be reset here.
@@ -772,6 +815,12 @@ pub fn apply_settings_to_builder(
Setting::Reset => builder.reset_facet_search(),
Setting::NotSet => (),
}
match chat {
Setting::Set(chat) => builder.set_chat(chat.clone()),
Setting::Reset => builder.reset_chat(),
Setting::NotSet => (),
}
}
pub enum SecretPolicy {
@@ -826,12 +875,14 @@ pub fn settings(
};
let disabled_attributes = index.exact_attributes(rtxn)?.into_iter().map(String::from).collect();
let DisabledTyposTerms { disable_on_numbers } = index.disabled_typos_terms(rtxn)?;
let typo_tolerance = TypoSettings {
enabled: Setting::Set(index.authorize_typos(rtxn)?),
min_word_size_for_typos: Setting::Set(min_typo_word_len),
disable_on_words: Setting::Set(disabled_words),
disable_on_attributes: Setting::Set(disabled_attributes),
disable_on_numbers: Setting::Set(disable_on_numbers),
};
let faceting = FacetingSettings {
@@ -867,14 +918,11 @@ pub fn settings(
})
.collect();
let embedders = Setting::Set(embedders);
let search_cutoff_ms = index.search_cutoff(rtxn)?;
let localized_attributes_rules = index.localized_attributes_rules(rtxn)?;
let prefix_search = index.prefix_search(rtxn)?.map(PrefixSearchSettings::from);
let facet_search = index.facet_search(rtxn)?;
let chat = index.chat_config(rtxn).map(ChatSettings::from)?;
let mut settings = Settings {
displayed_attributes: match displayed_attributes {
@@ -912,8 +960,9 @@ pub fn settings(
Some(rules) => Setting::Set(rules.into_iter().map(|r| r.into()).collect()),
None => Setting::Reset,
},
prefix_search: Setting::Set(prefix_search.unwrap_or_default()),
facet_search: Setting::Set(facet_search),
prefix_search: Setting::Set(prefix_search.unwrap_or_default()),
chat: Setting::Set(chat),
_kind: PhantomData,
};
@@ -1141,6 +1190,7 @@ pub(crate) mod test {
search_cutoff_ms: Setting::NotSet,
facet_search: Setting::NotSet,
prefix_search: Setting::NotSet,
chat: Setting::NotSet,
_kind: PhantomData::<Unchecked>,
};
@@ -1172,6 +1222,8 @@ pub(crate) mod test {
search_cutoff_ms: Setting::NotSet,
facet_search: Setting::NotSet,
prefix_search: Setting::NotSet,
chat: Setting::NotSet,
_kind: PhantomData::<Unchecked>,
};

View File

@@ -272,9 +272,9 @@ impl KindWithContent {
KindWithContent::UpgradeDatabase { from } => Some(Details::UpgradeDatabase {
from: (from.0, from.1, from.2),
to: (
versioning::VERSION_MAJOR.parse().unwrap(),
versioning::VERSION_MINOR.parse().unwrap(),
versioning::VERSION_PATCH.parse().unwrap(),
versioning::VERSION_MAJOR,
versioning::VERSION_MINOR,
versioning::VERSION_PATCH,
),
}),
}
@@ -338,9 +338,9 @@ impl KindWithContent {
KindWithContent::UpgradeDatabase { from } => Some(Details::UpgradeDatabase {
from: *from,
to: (
versioning::VERSION_MAJOR.parse().unwrap(),
versioning::VERSION_MINOR.parse().unwrap(),
versioning::VERSION_PATCH.parse().unwrap(),
versioning::VERSION_MAJOR,
versioning::VERSION_MINOR,
versioning::VERSION_PATCH,
),
}),
}
@@ -386,9 +386,9 @@ impl From<&KindWithContent> for Option<Details> {
KindWithContent::UpgradeDatabase { from } => Some(Details::UpgradeDatabase {
from: *from,
to: (
versioning::VERSION_MAJOR.parse().unwrap(),
versioning::VERSION_MINOR.parse().unwrap(),
versioning::VERSION_PATCH.parse().unwrap(),
versioning::VERSION_MAJOR,
versioning::VERSION_MINOR,
versioning::VERSION_PATCH,
),
}),
}

View File

@@ -8,9 +8,7 @@ use tempfile::NamedTempFile;
/// The name of the file that contains the version of the database.
pub const VERSION_FILE_NAME: &str = "VERSION";
pub static VERSION_MAJOR: &str = env!("CARGO_PKG_VERSION_MAJOR");
pub static VERSION_MINOR: &str = env!("CARGO_PKG_VERSION_MINOR");
pub static VERSION_PATCH: &str = env!("CARGO_PKG_VERSION_PATCH");
pub use milli::constants::{VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH};
/// Persists the version of the current Meilisearch binary to a VERSION file
pub fn create_current_version_file(db_path: &Path) -> anyhow::Result<()> {
@@ -19,9 +17,9 @@ pub fn create_current_version_file(db_path: &Path) -> anyhow::Result<()> {
pub fn create_version_file(
db_path: &Path,
major: &str,
minor: &str,
patch: &str,
major: u32,
minor: u32,
patch: u32,
) -> anyhow::Result<()> {
let version_path = db_path.join(VERSION_FILE_NAME);
// In order to persist the file later we must create it in the `data.ms` and not in `/tmp`

View File

@@ -32,6 +32,7 @@ async-trait = "0.1.85"
bstr = "1.11.3"
byte-unit = { version = "5.1.6", features = ["serde"] }
bytes = "1.9.0"
bumpalo = "3.16.0"
clap = { version = "4.5.24", features = ["derive", "env"] }
crossbeam-channel = "0.5.15"
deserr = { version = "0.6.3", features = ["actix-web"] }
@@ -48,6 +49,7 @@ is-terminal = "0.4.13"
itertools = "0.14.0"
jsonwebtoken = "9.3.0"
lazy_static = "1.5.0"
liquid = "0.26.9"
meilisearch-auth = { path = "../meilisearch-auth" }
meilisearch-types = { path = "../meilisearch-types" }
mimalloc = { version = "0.1.43", default-features = false }
@@ -111,6 +113,8 @@ utoipa = { version = "5.3.1", features = [
"openapi_extensions",
] }
utoipa-scalar = { version = "0.3.0", optional = true, features = ["actix-web"] }
async-openai = { git = "https://github.com/meilisearch/async-openai", branch = "optional-type-function" }
actix-web-lab = { version = "0.24.1", default-features = false }
[dev-dependencies]
actix-rt = "2.10.0"

View File

@@ -4,6 +4,7 @@ use std::marker::PhantomData;
use std::ops::Deref;
use std::pin::Pin;
use actix_web::http::header::AUTHORIZATION;
use actix_web::web::Data;
use actix_web::FromRequest;
pub use error::AuthenticationError;
@@ -94,36 +95,44 @@ impl<P: Policy + 'static, D: 'static + Clone> FromRequest for GuardedData<P, D>
_payload: &mut actix_web::dev::Payload,
) -> Self::Future {
match req.app_data::<Data<AuthController>>().cloned() {
Some(auth) => match req
.headers()
.get("Authorization")
.map(|type_token| type_token.to_str().unwrap_or_default().splitn(2, ' '))
{
Some(mut type_token) => match type_token.next() {
Some("Bearer") => {
// TODO: find a less hardcoded way?
let index = req.match_info().get("index_uid");
match type_token.next() {
Some(token) => Box::pin(Self::auth_bearer(
auth,
token.to_string(),
index.map(String::from),
req.app_data::<D>().cloned(),
)),
None => Box::pin(err(AuthenticationError::InvalidToken.into())),
}
}
_otherwise => {
Box::pin(err(AuthenticationError::MissingAuthorizationHeader.into()))
}
},
None => Box::pin(Self::auth_token(auth, req.app_data::<D>().cloned())),
Some(auth) => match extract_token_from_request(req) {
Ok(Some(token)) => {
// TODO: find a less hardcoded way?
let index = req.match_info().get("index_uid");
Box::pin(Self::auth_bearer(
auth,
token.to_string(),
index.map(String::from),
req.app_data::<D>().cloned(),
))
}
Ok(None) => Box::pin(Self::auth_token(auth, req.app_data::<D>().cloned())),
Err(e) => Box::pin(err(e.into())),
},
None => Box::pin(err(AuthenticationError::IrretrievableState.into())),
}
}
}
pub fn extract_token_from_request(
req: &actix_web::HttpRequest,
) -> Result<Option<&str>, AuthenticationError> {
match req
.headers()
.get(AUTHORIZATION)
.map(|type_token| type_token.to_str().unwrap_or_default().splitn(2, ' '))
{
Some(mut type_token) => match type_token.next() {
Some("Bearer") => match type_token.next() {
Some(token) => Ok(Some(token)),
None => Err(AuthenticationError::InvalidToken),
},
_otherwise => Err(AuthenticationError::MissingAuthorizationHeader),
},
None => Ok(None),
}
}
pub trait Policy {
fn authenticate(
auth: Data<AuthController>,
@@ -299,8 +308,8 @@ pub mod policies {
auth: &AuthController,
token: &str,
) -> Result<TenantTokenOutcome, AuthError> {
// Only search action can be accessed by a tenant token.
if A != actions::SEARCH {
// Only search and chat actions can be accessed by a tenant token.
if A != actions::SEARCH && A != actions::CHAT {
return Ok(TenantTokenOutcome::NotATenantToken);
}

View File

@@ -235,10 +235,7 @@ pub fn setup_meilisearch(opt: &Opt) -> anyhow::Result<(Arc<IndexScheduler>, Arc<
auto_upgrade: opt.experimental_dumpless_upgrade,
embedding_cache_cap: opt.experimental_embedding_cache_entries,
};
let bin_major: u32 = VERSION_MAJOR.parse().unwrap();
let bin_minor: u32 = VERSION_MINOR.parse().unwrap();
let bin_patch: u32 = VERSION_PATCH.parse().unwrap();
let binary_version = (bin_major, bin_minor, bin_patch);
let binary_version = (VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH);
let empty_db = is_empty_db(&opt.db_path);
let (index_scheduler, auth_controller) = if let Some(ref snapshot_path) = opt.import_snapshot {

View File

@@ -0,0 +1,560 @@
use std::cell::RefCell;
use std::collections::HashMap;
use std::mem;
use std::sync::RwLock;
use std::time::Duration;
use actix_web::web::{self, Data};
use actix_web::{Either, HttpRequest, HttpResponse, Responder};
use actix_web_lab::sse::{self, Event, Sse};
use async_openai::config::OpenAIConfig;
use async_openai::types::{
ChatCompletionMessageToolCall, ChatCompletionMessageToolCallChunk,
ChatCompletionRequestAssistantMessageArgs, ChatCompletionRequestMessage,
ChatCompletionRequestSystemMessage, ChatCompletionRequestSystemMessageContent,
ChatCompletionRequestToolMessage, ChatCompletionRequestToolMessageContent,
ChatCompletionStreamResponseDelta, ChatCompletionToolArgs, ChatCompletionToolType,
CreateChatCompletionRequest, FinishReason, FunctionCall, FunctionCallStream,
FunctionObjectArgs,
};
use async_openai::Client;
use bumpalo::Bump;
use futures::StreamExt;
use index_scheduler::IndexScheduler;
use meilisearch_auth::AuthController;
use meilisearch_types::error::ResponseError;
use meilisearch_types::heed::RoTxn;
use meilisearch_types::keys::actions;
use meilisearch_types::milli::index::ChatConfig;
use meilisearch_types::milli::prompt::{Prompt, PromptData};
use meilisearch_types::milli::update::new::document::DocumentFromDb;
use meilisearch_types::milli::{
DocumentId, FieldIdMapWithMetadata, GlobalFieldsIdsMap, MetadataBuilder, TimeBudget,
};
use meilisearch_types::Index;
use serde::Deserialize;
use serde_json::json;
use tokio::runtime::Handle;
use tokio::sync::mpsc::error::SendError;
use super::settings::chat::{ChatPrompts, GlobalChatSettings};
use crate::error::MeilisearchHttpError;
use crate::extractors::authentication::policies::ActionPolicy;
use crate::extractors::authentication::{extract_token_from_request, GuardedData, Policy as _};
use crate::metrics::MEILISEARCH_DEGRADED_SEARCH_REQUESTS;
use crate::routes::indexes::search::search_kind;
use crate::search::{
add_search_rules, prepare_search, search_from_kind, HybridQuery, MatchingStrategy, SearchQuery,
SemanticRatio,
};
use crate::search_queue::SearchQueue;
const EMBEDDER_NAME: &str = "openai";
const SEARCH_IN_INDEX_FUNCTION_NAME: &str = "_meiliSearchInIndex";
pub fn configure(cfg: &mut web::ServiceConfig) {
cfg.service(web::resource("/completions").route(web::post().to(chat)));
}
/// Get a chat completion
async fn chat(
index_scheduler: GuardedData<ActionPolicy<{ actions::CHAT }>, Data<IndexScheduler>>,
auth_ctrl: web::Data<AuthController>,
req: HttpRequest,
search_queue: web::Data<SearchQueue>,
web::Json(chat_completion): web::Json<CreateChatCompletionRequest>,
) -> impl Responder {
// To enable later on, when the feature will be experimental
// index_scheduler.features().check_chat("Using the /chat route")?;
assert_eq!(
chat_completion.n.unwrap_or(1),
1,
"Meilisearch /chat only support one completion at a time (n = 1, n = null)"
);
if chat_completion.stream.unwrap_or(false) {
Either::Right(
streamed_chat(index_scheduler, auth_ctrl, req, search_queue, chat_completion).await,
)
} else {
Either::Left(
non_streamed_chat(index_scheduler, auth_ctrl, req, search_queue, chat_completion).await,
)
}
}
/// Setup search tool in chat completion request
fn setup_search_tool(
index_scheduler: &Data<IndexScheduler>,
filters: &meilisearch_auth::AuthFilter,
chat_completion: &mut CreateChatCompletionRequest,
prompts: &ChatPrompts,
) -> Result<(), ResponseError> {
let tools = chat_completion.tools.get_or_insert_default();
if tools.iter().find(|t| t.function.name == SEARCH_IN_INDEX_FUNCTION_NAME).is_some() {
panic!("{SEARCH_IN_INDEX_FUNCTION_NAME} function already set");
}
let index_uids: Vec<_> = index_scheduler
.index_names()?
.into_iter()
.filter(|index_uid| filters.is_index_authorized(&index_uid))
.collect();
let tool = ChatCompletionToolArgs::default()
.r#type(ChatCompletionToolType::Function)
.function(
FunctionObjectArgs::default()
.name(SEARCH_IN_INDEX_FUNCTION_NAME)
.description(&prompts.search_description)
.parameters(json!({
"type": "object",
"properties": {
"index_uid": {
"type": "string",
"enum": index_uids,
"description": prompts.search_index_uid_param,
},
"q": {
// Unfortunately, Mistral does not support an array of types, here.
// "type": ["string", "null"],
"type": "string",
"description": prompts.search_q_param,
}
},
"required": ["index_uid", "q"],
"additionalProperties": false,
}))
.strict(true)
.build()
.unwrap(),
)
.build()
.unwrap();
tools.push(tool);
chat_completion.messages.insert(
0,
ChatCompletionRequestMessage::System(ChatCompletionRequestSystemMessage {
content: ChatCompletionRequestSystemMessageContent::Text(prompts.system.clone()),
name: None,
}),
);
Ok(())
}
/// Process search request and return formatted results
async fn process_search_request(
index_scheduler: &GuardedData<ActionPolicy<{ actions::CHAT }>, Data<IndexScheduler>>,
auth_ctrl: web::Data<AuthController>,
search_queue: &web::Data<SearchQueue>,
auth_token: &str,
index_uid: String,
q: Option<String>,
) -> Result<(Index, String), ResponseError> {
let mut query = SearchQuery {
q,
hybrid: Some(HybridQuery {
semantic_ratio: SemanticRatio::default(),
embedder: EMBEDDER_NAME.to_string(),
}),
limit: 20,
matching_strategy: MatchingStrategy::Frequency,
..Default::default()
};
let auth_filter = ActionPolicy::<{ actions::SEARCH }>::authenticate(
auth_ctrl,
auth_token,
Some(index_uid.as_str()),
)?;
// Tenant token search_rules.
if let Some(search_rules) = auth_filter.get_index_search_rules(&index_uid) {
add_search_rules(&mut query.filter, search_rules);
}
// TBD
// let mut aggregate = SearchAggregator::<SearchPOST>::from_query(&query);
let index = index_scheduler.index(&index_uid)?;
let search_kind =
search_kind(&query, index_scheduler.get_ref(), index_uid.to_string(), &index)?;
let permit = search_queue.try_get_search_permit().await?;
let features = index_scheduler.features();
let index_cloned = index.clone();
let search_result = tokio::task::spawn_blocking(move || -> Result<_, ResponseError> {
let rtxn = index_cloned.read_txn()?;
let time_budget = match index_cloned
.search_cutoff(&rtxn)
.map_err(|e| MeilisearchHttpError::from_milli(e, Some(index_uid.clone())))?
{
Some(cutoff) => TimeBudget::new(Duration::from_millis(cutoff)),
None => TimeBudget::default(),
};
let (search, _is_finite_pagination, _max_total_hits, _offset) =
prepare_search(&index_cloned, &rtxn, &query, &search_kind, time_budget, features)?;
search_from_kind(index_uid, search_kind, search)
.map(|(search_results, _)| search_results)
.map_err(ResponseError::from)
})
.await;
permit.drop().await;
let search_result = search_result?;
if let Ok(ref search_result) = search_result {
// aggregate.succeed(search_result);
if search_result.degraded {
MEILISEARCH_DEGRADED_SEARCH_REQUESTS.inc();
}
}
// analytics.publish(aggregate, &req);
let search_result = search_result?;
let rtxn = index.read_txn()?;
let render_alloc = Bump::new();
let formatted = format_documents(&rtxn, &index, &render_alloc, search_result.documents_ids)?;
let text = formatted.join("\n");
drop(rtxn);
Ok((index, text))
}
async fn non_streamed_chat(
index_scheduler: GuardedData<ActionPolicy<{ actions::CHAT }>, Data<IndexScheduler>>,
auth_ctrl: web::Data<AuthController>,
req: HttpRequest,
search_queue: web::Data<SearchQueue>,
mut chat_completion: CreateChatCompletionRequest,
) -> Result<HttpResponse, ResponseError> {
let filters = index_scheduler.filters();
let chat_settings = match index_scheduler.chat_settings().unwrap() {
Some(value) => serde_json::from_value(value).unwrap(),
None => GlobalChatSettings::default(),
};
let mut config = OpenAIConfig::default();
if let Some(api_key) = chat_settings.api_key.as_ref() {
config = config.with_api_key(api_key);
}
if let Some(base_api) = chat_settings.base_api.as_ref() {
config = config.with_api_base(base_api);
}
let client = Client::with_config(config);
let auth_token = extract_token_from_request(&req)?.unwrap();
setup_search_tool(&index_scheduler, filters, &mut chat_completion, &chat_settings.prompts)?;
let mut response;
loop {
response = client.chat().create(chat_completion.clone()).await.unwrap();
let choice = &mut response.choices[0];
match choice.finish_reason {
Some(FinishReason::ToolCalls) => {
let tool_calls = mem::take(&mut choice.message.tool_calls).unwrap_or_default();
let (meili_calls, other_calls): (Vec<_>, Vec<_>) = tool_calls
.into_iter()
.partition(|call| call.function.name == SEARCH_IN_INDEX_FUNCTION_NAME);
chat_completion.messages.push(
ChatCompletionRequestAssistantMessageArgs::default()
.tool_calls(meili_calls.clone())
.build()
.unwrap()
.into(),
);
for call in meili_calls {
let result = match serde_json::from_str(&call.function.arguments) {
Ok(SearchInIndexParameters { index_uid, q }) => process_search_request(
&index_scheduler,
auth_ctrl.clone(),
&search_queue,
&auth_token,
index_uid,
q,
)
.await
.map_err(|e| e.to_string()),
Err(err) => Err(err.to_string()),
};
let text = match result {
Ok((_, text)) => text,
Err(err) => err,
};
chat_completion.messages.push(ChatCompletionRequestMessage::Tool(
ChatCompletionRequestToolMessage {
tool_call_id: call.id.clone(),
content: ChatCompletionRequestToolMessageContent::Text(format!(
"{}\n\n{text}",
chat_settings.prompts.pre_query
)),
},
));
}
// Let the client call other tools by themselves
if !other_calls.is_empty() {
response.choices[0].message.tool_calls = Some(other_calls);
break;
}
}
_ => break,
}
}
Ok(HttpResponse::Ok().json(response))
}
async fn streamed_chat(
index_scheduler: GuardedData<ActionPolicy<{ actions::CHAT }>, Data<IndexScheduler>>,
auth_ctrl: web::Data<AuthController>,
req: HttpRequest,
search_queue: web::Data<SearchQueue>,
mut chat_completion: CreateChatCompletionRequest,
) -> Result<impl Responder, ResponseError> {
let filters = index_scheduler.filters();
let chat_settings = match index_scheduler.chat_settings().unwrap() {
Some(value) => serde_json::from_value(value).unwrap(),
None => GlobalChatSettings::default(),
};
let mut config = OpenAIConfig::default();
if let Some(api_key) = chat_settings.api_key.as_ref() {
config = config.with_api_key(api_key);
}
if let Some(base_api) = chat_settings.base_api.as_ref() {
config = config.with_api_base(base_api);
}
let auth_token = extract_token_from_request(&req)?.unwrap().to_string();
setup_search_tool(&index_scheduler, filters, &mut chat_completion, &chat_settings.prompts)?;
let (tx, rx) = tokio::sync::mpsc::channel(10);
let _join_handle = Handle::current().spawn(async move {
let client = Client::with_config(config.clone());
let mut global_tool_calls = HashMap::<u32, Call>::new();
let mut finish_reason = None;
// Limit the number of internal calls to satisfy the search requests of the LLM
'main: for _ in 0..20 {
let mut response = client.chat().create_stream(chat_completion.clone()).await.unwrap();
while let Some(result) = response.next().await {
match result {
Ok(resp) => {
let choice = &resp.choices[0];
finish_reason = choice.finish_reason;
#[allow(deprecated)]
let ChatCompletionStreamResponseDelta {
content,
// Using deprecated field but keeping for compatibility
function_call: _,
ref tool_calls,
role: _,
refusal: _,
} = &choice.delta;
if content.is_some() {
if let Err(SendError(_)) = tx.send(Event::Data(sse::Data::new_json(&resp).unwrap())).await {
return;
}
}
match tool_calls {
Some(tool_calls) => {
for chunk in tool_calls {
let ChatCompletionMessageToolCallChunk {
index,
id,
r#type: _,
function,
} = chunk;
let FunctionCallStream { name, arguments } =
function.as_ref().unwrap();
global_tool_calls
.entry(*index)
.and_modify(|call| call.append(arguments.as_ref().unwrap()))
.or_insert_with(|| Call {
id: id.as_ref().unwrap().clone(),
function_name: name.as_ref().unwrap().clone(),
arguments: arguments.as_ref().unwrap().clone(),
});
}
}
None if !global_tool_calls.is_empty() => {
let (meili_calls, _other_calls): (Vec<_>, Vec<_>) =
mem::take(&mut global_tool_calls)
.into_values()
.map(|call| ChatCompletionMessageToolCall {
id: call.id,
r#type: Some(ChatCompletionToolType::Function),
function: FunctionCall {
name: call.function_name,
arguments: call.arguments,
},
})
.partition(|call| call.function.name == SEARCH_IN_INDEX_FUNCTION_NAME);
chat_completion.messages.push(
ChatCompletionRequestAssistantMessageArgs::default()
.tool_calls(meili_calls.clone())
.build()
.unwrap()
.into(),
);
for call in meili_calls {
if let Err(SendError(_)) = tx.send(Event::Data(
sse::Data::new_json(json!({
"object": "chat.completion.tool.call",
"tool": call,
}))
.unwrap(),
))
.await {
return;
}
let result = match serde_json::from_str(&call.function.arguments) {
Ok(SearchInIndexParameters { index_uid, q }) => process_search_request(
&index_scheduler,
auth_ctrl.clone(),
&search_queue,
&auth_token,
index_uid,
q,
).await.map_err(|e| e.to_string()),
Err(err) => Err(err.to_string()),
};
let is_error = result.is_err();
let text = match result {
Ok((_, text)) => text,
Err(err) => err,
};
let tool = ChatCompletionRequestToolMessage {
tool_call_id: call.id.clone(),
content: ChatCompletionRequestToolMessageContent::Text(
format!("{}\n\n{text}", chat_settings.prompts.pre_query),
),
};
if let Err(SendError(_)) = tx.send(Event::Data(
sse::Data::new_json(json!({
"object": if is_error {
"chat.completion.tool.error"
} else {
"chat.completion.tool.output"
},
"tool": ChatCompletionRequestToolMessage {
tool_call_id: call.id,
content: ChatCompletionRequestToolMessageContent::Text(
text,
),
},
}))
.unwrap(),
))
.await {
return;
}
chat_completion.messages.push(ChatCompletionRequestMessage::Tool(tool));
}
}
None => (),
}
}
Err(err) => {
tracing::error!("{err:?}");
if let Err(SendError(_)) = tx.send(Event::Data(sse::Data::new_json(&json!({
"object": "chat.completion.error",
"tool": err.to_string(),
})).unwrap())).await {
return;
}
break 'main;
}
}
}
// We must stop if the finish reason is not something we can solve with Meilisearch
if finish_reason.map_or(true, |fr| fr != FinishReason::ToolCalls) {
break;
}
}
let _ = tx.send(Event::Data(sse::Data::new("[DONE]")));
});
Ok(Sse::from_infallible_receiver(rx).with_retry_duration(Duration::from_secs(10)))
}
/// The structure used to aggregate the function calls to make.
#[derive(Debug)]
struct Call {
id: String,
function_name: String,
arguments: String,
}
impl Call {
fn append(&mut self, arguments: &str) {
self.arguments.push_str(arguments);
}
}
#[derive(Deserialize)]
struct SearchInIndexParameters {
/// The index uid to search in.
index_uid: String,
/// The query parameter to use.
q: Option<String>,
}
fn format_documents<'t, 'doc>(
rtxn: &RoTxn<'t>,
index: &Index,
doc_alloc: &'doc Bump,
internal_docids: Vec<DocumentId>,
) -> Result<Vec<&'doc str>, ResponseError> {
let ChatConfig { prompt: PromptData { template, max_bytes }, .. } = index.chat_config(rtxn)?;
let prompt = Prompt::new(template, max_bytes).unwrap();
let fid_map = index.fields_ids_map(rtxn)?;
let metadata_builder = MetadataBuilder::from_index(index, rtxn)?;
let fid_map_with_meta = FieldIdMapWithMetadata::new(fid_map.clone(), metadata_builder);
let global = RwLock::new(fid_map_with_meta);
let gfid_map = RefCell::new(GlobalFieldsIdsMap::new(&global));
let external_ids: Vec<String> = index
.external_id_of(rtxn, internal_docids.iter().copied())?
.into_iter()
.collect::<Result<_, _>>()?;
let mut renders = Vec::new();
for (docid, external_docid) in internal_docids.into_iter().zip(external_ids) {
let document = match DocumentFromDb::new(docid, rtxn, index, &fid_map)? {
Some(doc) => doc,
None => continue,
};
let text = prompt.render_document(&external_docid, document, &gfid_map, doc_alloc).unwrap();
renders.push(text);
}
Ok(renders)
}

View File

@@ -6,7 +6,7 @@ use meilisearch_types::deserr::DeserrJsonError;
use meilisearch_types::error::ResponseError;
use meilisearch_types::index_uid::IndexUid;
use meilisearch_types::settings::{
settings, SecretPolicy, SettingEmbeddingSettings, Settings, Unchecked,
settings, ChatSettings, SecretPolicy, SettingEmbeddingSettings, Settings, Unchecked,
};
use meilisearch_types::tasks::KindWithContent;
use tracing::debug;
@@ -508,6 +508,17 @@ make_setting_routes!(
camelcase_attr: "prefixSearch",
analytics: PrefixSearchAnalytics
},
{
route: "/chat",
update_verb: put,
value_type: ChatSettings,
err_type: meilisearch_types::deserr::DeserrJsonError<
meilisearch_types::error::deserr_codes::InvalidSettingsIndexChat,
>,
attr: chat,
camelcase_attr: "chat",
analytics: ChatAnalytics
},
);
#[utoipa::path(
@@ -597,6 +608,7 @@ pub async fn update_all(
),
facet_search: FacetSearchAnalytics::new(new_settings.facet_search.as_ref().set()),
prefix_search: PrefixSearchAnalytics::new(new_settings.prefix_search.as_ref().set()),
chat: ChatAnalytics::new(new_settings.chat.as_ref().set()),
},
&req,
);

View File

@@ -10,8 +10,8 @@ use meilisearch_types::locales::{Locale, LocalizedAttributesRuleView};
use meilisearch_types::milli::update::Setting;
use meilisearch_types::milli::FilterableAttributesRule;
use meilisearch_types::settings::{
FacetingSettings, PaginationSettings, PrefixSearchSettings, ProximityPrecisionView,
RankingRuleView, SettingEmbeddingSettings, TypoSettings,
ChatSettings, FacetingSettings, PaginationSettings, PrefixSearchSettings,
ProximityPrecisionView, RankingRuleView, SettingEmbeddingSettings, TypoSettings,
};
use serde::Serialize;
@@ -39,6 +39,7 @@ pub struct SettingsAnalytics {
pub non_separator_tokens: NonSeparatorTokensAnalytics,
pub facet_search: FacetSearchAnalytics,
pub prefix_search: PrefixSearchAnalytics,
pub chat: ChatAnalytics,
}
impl Aggregate for SettingsAnalytics {
@@ -198,6 +199,7 @@ impl Aggregate for SettingsAnalytics {
set: new.prefix_search.set | self.prefix_search.set,
value: new.prefix_search.value.or(self.prefix_search.value),
},
chat: ChatAnalytics { set: new.chat.set | self.chat.set },
})
}
@@ -674,3 +676,18 @@ impl PrefixSearchAnalytics {
SettingsAnalytics { prefix_search: self, ..Default::default() }
}
}
#[derive(Serialize, Default)]
pub struct ChatAnalytics {
pub set: bool,
}
impl ChatAnalytics {
pub fn new(settings: Option<&ChatSettings>) -> Self {
Self { set: settings.is_some() }
}
pub fn into_settings(self) -> SettingsAnalytics {
SettingsAnalytics { chat: self, ..Default::default() }
}
}

View File

@@ -52,6 +52,7 @@ const PAGINATION_DEFAULT_LIMIT_FN: fn() -> usize = || 20;
mod api_key;
pub mod batches;
pub mod chat;
mod dump;
pub mod features;
pub mod indexes;
@@ -61,6 +62,7 @@ mod multi_search;
mod multi_search_analytics;
pub mod network;
mod open_api_utils;
pub mod settings;
mod snapshot;
mod swap_indexes;
pub mod tasks;
@@ -113,7 +115,9 @@ pub fn configure(cfg: &mut web::ServiceConfig) {
.service(web::scope("/swap-indexes").configure(swap_indexes::configure))
.service(web::scope("/metrics").configure(metrics::configure))
.service(web::scope("/experimental-features").configure(features::configure))
.service(web::scope("/network").configure(network::configure));
.service(web::scope("/network").configure(network::configure))
.service(web::scope("/chat").configure(chat::configure))
.service(web::scope("/settings/chat").configure(settings::chat::configure));
#[cfg(feature = "swagger")]
{

View File

@@ -0,0 +1,107 @@
use actix_web::web::{self, Data};
use actix_web::HttpResponse;
use index_scheduler::IndexScheduler;
use meilisearch_types::error::ResponseError;
use meilisearch_types::keys::actions;
use serde::{Deserialize, Serialize};
use crate::extractors::authentication::policies::ActionPolicy;
use crate::extractors::authentication::GuardedData;
use crate::extractors::sequential_extractor::SeqHandler;
pub fn configure(cfg: &mut web::ServiceConfig) {
cfg.service(
web::resource("")
.route(web::get().to(get_settings))
.route(web::patch().to(SeqHandler(patch_settings))),
);
}
async fn get_settings(
index_scheduler: GuardedData<
ActionPolicy<{ actions::CHAT_SETTINGS_GET }>,
Data<IndexScheduler>,
>,
) -> Result<HttpResponse, ResponseError> {
let settings = match index_scheduler.chat_settings()? {
Some(value) => serde_json::from_value(value).unwrap(),
None => GlobalChatSettings::default(),
};
Ok(HttpResponse::Ok().json(settings))
}
async fn patch_settings(
index_scheduler: GuardedData<
ActionPolicy<{ actions::CHAT_SETTINGS_UPDATE }>,
Data<IndexScheduler>,
>,
web::Json(chat_settings): web::Json<GlobalChatSettings>,
) -> Result<HttpResponse, ResponseError> {
let chat_settings = serde_json::to_value(chat_settings).unwrap();
index_scheduler.put_chat_settings(&chat_settings)?;
Ok(HttpResponse::Ok().finish())
}
#[derive(Debug, Serialize, Deserialize)]
#[serde(deny_unknown_fields, rename_all = "camelCase")]
pub struct GlobalChatSettings {
pub source: String,
pub base_api: Option<String>,
pub api_key: Option<String>,
pub prompts: ChatPrompts,
}
#[derive(Debug, Serialize, Deserialize)]
#[serde(deny_unknown_fields, rename_all = "camelCase")]
pub struct ChatPrompts {
pub system: String,
pub search_description: String,
pub search_q_param: String,
pub search_index_uid_param: String,
pub pre_query: String,
}
#[derive(Debug, Serialize, Deserialize)]
#[serde(deny_unknown_fields, rename_all = "camelCase")]
pub struct ChatIndexSettings {
pub description: String,
pub document_template: String,
}
const DEFAULT_SYSTEM_MESSAGE: &str = "You are a highly capable research assistant with access to powerful search tools. IMPORTANT INSTRUCTIONS:\
1. When answering questions, you MUST make multiple tool calls (at least 2-3) to gather comprehensive information.\
2. Use different search queries for each tool call - vary keywords, rephrase questions, and explore different semantic angles to ensure broad coverage.\
3. Always explicitly announce BEFORE making each tool call by saying: \"I'll search for [specific information] now.\"\
4. Combine information from ALL tool calls to provide complete, nuanced answers rather than relying on a single source.\
5. For complex topics, break down your research into multiple targeted queries rather than using a single generic search.";
/// The default description of the searchInIndex tool provided to OpenAI.
const DEFAULT_SEARCH_IN_INDEX_TOOL_DESCRIPTION: &str =
"Search the database for relevant JSON documents using an optional query.";
/// The default description of the searchInIndex `q` parameter tool provided to OpenAI.
const DEFAULT_SEARCH_IN_INDEX_Q_PARAMETER_TOOL_DESCRIPTION: &str =
"The search query string used to find relevant documents in the index. \
This should contain keywords or phrases that best represent what the user is looking for. \
More specific queries will yield more precise results.";
/// The default description of the searchInIndex `index` parameter tool provided to OpenAI.
const DEFAULT_SEARCH_IN_INDEX_INDEX_PARAMETER_TOOL_DESCRIPTION: &str =
"The name of the index to search within. An index is a collection of documents organized for search. \
Selecting the right index ensures the most relevant results for the user query";
impl Default for GlobalChatSettings {
fn default() -> Self {
GlobalChatSettings {
source: "openAi".to_string(),
base_api: None,
api_key: None,
prompts: ChatPrompts {
system: DEFAULT_SYSTEM_MESSAGE.to_string(),
search_description: DEFAULT_SEARCH_IN_INDEX_TOOL_DESCRIPTION.to_string(),
search_q_param: DEFAULT_SEARCH_IN_INDEX_Q_PARAMETER_TOOL_DESCRIPTION.to_string(),
search_index_uid_param: DEFAULT_SEARCH_IN_INDEX_INDEX_PARAMETER_TOOL_DESCRIPTION
.to_string(),
pre_query: "".to_string(),
},
}
}
}

View File

@@ -0,0 +1 @@
pub mod chat;

View File

@@ -882,7 +882,7 @@ pub fn add_search_rules(filter: &mut Option<Value>, rules: IndexSearchRules) {
}
}
fn prepare_search<'t>(
pub fn prepare_search<'t>(
index: &'t Index,
rtxn: &'t RoTxn,
query: &'t SearchQuery,

View File

@@ -820,6 +820,22 @@ async fn list_api_keys() {
"createdAt": "[ignored]",
"updatedAt": "[ignored]"
},
{
"name": "Default Chat API Key",
"description": "Use it to chat and search from the frontend",
"key": "[ignored]",
"uid": "[ignored]",
"actions": [
"search",
"chat.get"
],
"indexes": [
"*"
],
"expiresAt": null,
"createdAt": "[ignored]",
"updatedAt": "[ignored]"
},
{
"name": "Default Search API Key",
"description": "Use it to search from the frontend",

View File

@@ -310,7 +310,7 @@ async fn test_summarized_document_addition_or_update() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "batched all enqueued tasks"
"batchCreationComplete": "batched all enqueued tasks"
}
"###);
@@ -353,7 +353,7 @@ async fn test_summarized_document_addition_or_update() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "batched all enqueued tasks"
"batchCreationComplete": "batched all enqueued tasks"
}
"###);
}
@@ -398,7 +398,7 @@ async fn test_summarized_delete_documents_by_batch() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "batched all enqueued tasks"
"batchCreationComplete": "batched all enqueued tasks"
}
"###);
@@ -440,7 +440,7 @@ async fn test_summarized_delete_documents_by_batch() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "batched all enqueued tasks"
"batchCreationComplete": "batched all enqueued tasks"
}
"###);
}
@@ -488,7 +488,7 @@ async fn test_summarized_delete_documents_by_filter() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "batched all enqueued tasks"
"batchCreationComplete": "batched all enqueued tasks"
}
"###);
@@ -532,7 +532,7 @@ async fn test_summarized_delete_documents_by_filter() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "batched all enqueued tasks"
"batchCreationComplete": "batched all enqueued tasks"
}
"###);
@@ -576,7 +576,7 @@ async fn test_summarized_delete_documents_by_filter() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "batched all enqueued tasks"
"batchCreationComplete": "batched all enqueued tasks"
}
"###);
}
@@ -622,7 +622,7 @@ async fn test_summarized_delete_document_by_id() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "batched all enqueued tasks"
"batchCreationComplete": "batched all enqueued tasks"
}
"###);
@@ -664,7 +664,7 @@ async fn test_summarized_delete_document_by_id() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "batched all enqueued tasks"
"batchCreationComplete": "batched all enqueued tasks"
}
"###);
}
@@ -731,7 +731,7 @@ async fn test_summarized_settings_update() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "batched all enqueued tasks"
"batchCreationComplete": "batched all enqueued tasks"
}
"###);
}
@@ -773,7 +773,7 @@ async fn test_summarized_index_creation() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "task with id 0 of type `indexCreation` cannot be batched"
"batchCreationComplete": "task with id 0 of type `indexCreation` cannot be batched"
}
"###);
@@ -812,7 +812,7 @@ async fn test_summarized_index_creation() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "task with id 1 of type `indexCreation` cannot be batched"
"batchCreationComplete": "task with id 1 of type `indexCreation` cannot be batched"
}
"###);
}
@@ -964,7 +964,7 @@ async fn test_summarized_index_update() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "task with id 0 of type `indexUpdate` cannot be batched"
"batchCreationComplete": "task with id 0 of type `indexUpdate` cannot be batched"
}
"###);
@@ -1003,7 +1003,7 @@ async fn test_summarized_index_update() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "task with id 1 of type `indexUpdate` cannot be batched"
"batchCreationComplete": "task with id 1 of type `indexUpdate` cannot be batched"
}
"###);
@@ -1043,7 +1043,7 @@ async fn test_summarized_index_update() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "task with id 3 of type `indexUpdate` cannot be batched"
"batchCreationComplete": "task with id 3 of type `indexUpdate` cannot be batched"
}
"###);
@@ -1082,7 +1082,7 @@ async fn test_summarized_index_update() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "task with id 4 of type `indexUpdate` cannot be batched"
"batchCreationComplete": "task with id 4 of type `indexUpdate` cannot be batched"
}
"###);
}
@@ -1134,7 +1134,7 @@ async fn test_summarized_index_swap() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "task with id 0 of type `indexSwap` cannot be batched"
"batchCreationComplete": "task with id 0 of type `indexSwap` cannot be batched"
}
"###);
@@ -1177,7 +1177,7 @@ async fn test_summarized_index_swap() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "task with id 1 of type `indexCreation` cannot be batched"
"batchCreationComplete": "task with id 1 of type `indexCreation` cannot be batched"
}
"###);
}
@@ -1224,7 +1224,7 @@ async fn test_summarized_batch_cancelation() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "task with id 1 of type `taskCancelation` cannot be batched"
"batchCreationComplete": "task with id 1 of type `taskCancelation` cannot be batched"
}
"###);
}
@@ -1271,7 +1271,7 @@ async fn test_summarized_batch_deletion() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "a batch of tasks of type `taskDeletion` cannot be batched with any other type of task"
"batchCreationComplete": "a batch of tasks of type `taskDeletion` cannot be batched with any other type of task"
}
"###);
}
@@ -1313,7 +1313,7 @@ async fn test_summarized_dump_creation() {
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "task with id 0 of type `dumpCreation` cannot be batched"
"batchCreationComplete": "task with id 0 of type `dumpCreation` cannot be batched"
}
"###);
}

View File

@@ -87,7 +87,8 @@ async fn import_dump_v1_movie_raw() {
"twoTypos": 9
},
"disableOnWords": [],
"disableOnAttributes": []
"disableOnAttributes": [],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 100,
@@ -260,7 +261,8 @@ async fn import_dump_v1_movie_with_settings() {
"twoTypos": 9
},
"disableOnWords": [],
"disableOnAttributes": []
"disableOnAttributes": [],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 100,
@@ -432,7 +434,8 @@ async fn import_dump_v1_rubygems_with_settings() {
"twoTypos": 9
},
"disableOnWords": [],
"disableOnAttributes": []
"disableOnAttributes": [],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 100,
@@ -590,7 +593,8 @@ async fn import_dump_v2_movie_raw() {
"twoTypos": 9
},
"disableOnWords": [],
"disableOnAttributes": []
"disableOnAttributes": [],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 100,
@@ -760,7 +764,8 @@ async fn import_dump_v2_movie_with_settings() {
"twoTypos": 9
},
"disableOnWords": [],
"disableOnAttributes": []
"disableOnAttributes": [],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 100,
@@ -929,7 +934,8 @@ async fn import_dump_v2_rubygems_with_settings() {
"twoTypos": 9
},
"disableOnWords": [],
"disableOnAttributes": []
"disableOnAttributes": [],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 100,
@@ -1087,7 +1093,8 @@ async fn import_dump_v3_movie_raw() {
"twoTypos": 9
},
"disableOnWords": [],
"disableOnAttributes": []
"disableOnAttributes": [],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 100,
@@ -1257,7 +1264,8 @@ async fn import_dump_v3_movie_with_settings() {
"twoTypos": 9
},
"disableOnWords": [],
"disableOnAttributes": []
"disableOnAttributes": [],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 100,
@@ -1426,7 +1434,8 @@ async fn import_dump_v3_rubygems_with_settings() {
"twoTypos": 9
},
"disableOnWords": [],
"disableOnAttributes": []
"disableOnAttributes": [],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 100,
@@ -1584,7 +1593,8 @@ async fn import_dump_v4_movie_raw() {
"twoTypos": 9
},
"disableOnWords": [],
"disableOnAttributes": []
"disableOnAttributes": [],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 100,
@@ -1754,7 +1764,8 @@ async fn import_dump_v4_movie_with_settings() {
"twoTypos": 9
},
"disableOnWords": [],
"disableOnAttributes": []
"disableOnAttributes": [],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 100,
@@ -1923,7 +1934,8 @@ async fn import_dump_v4_rubygems_with_settings() {
"twoTypos": 9
},
"disableOnWords": [],
"disableOnAttributes": []
"disableOnAttributes": [],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 100,
@@ -2212,7 +2224,8 @@ async fn import_dump_v6_containing_experimental_features() {
"twoTypos": 9
},
"disableOnWords": [],
"disableOnAttributes": []
"disableOnAttributes": [],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 100,
@@ -2444,7 +2457,8 @@ async fn generate_and_import_dump_containing_vectors() {
"twoTypos": 9
},
"disableOnWords": [],
"disableOnAttributes": []
"disableOnAttributes": [],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 100,

View File

@@ -27,7 +27,7 @@ source: crates/meilisearch/tests/dumps/mod.rs
"duration": "[date]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "batched all enqueued tasks"
"batchCreationComplete": "batched all enqueued tasks"
},
{
"uid": 1,
@@ -51,7 +51,7 @@ source: crates/meilisearch/tests/dumps/mod.rs
"duration": "PT0.144827890S",
"startedAt": "2025-02-04T10:15:21.275640274Z",
"finishedAt": "2025-02-04T10:15:21.420468164Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 0,
@@ -72,7 +72,7 @@ source: crates/meilisearch/tests/dumps/mod.rs
"duration": "PT0.032902186S",
"startedAt": "2025-02-04T10:14:43.559526162Z",
"finishedAt": "2025-02-04T10:14:43.592428348Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
}
],
"total": 3,

View File

@@ -1976,3 +1976,93 @@ async fn change_facet_casing() {
})
.await;
}
#[actix_rt::test]
async fn test_exact_typos_terms() {
let documents = json!([
{
"id": 0,
"title": "The zeroth document 1298484",
},
{
"id": 1,
"title": "The first document 234342",
"nested": {
"object": "field 22231",
"machin": "bidule 23443.32111",
},
},
{
"id": 2,
"title": "The second document 3398499",
"nested": [
"array",
{
"object": "field 23245121,23223",
},
{
"prout": "truc 123980612321",
"machin": "lol 12345645333447879",
},
],
},
{
"id": 3,
"title": "The third document 12333",
"nested": "I lied 98878",
},
]);
// Test prefix search
test_settings_documents_indexing_swapping_and_search(
&documents,
&json!({
"searchableAttributes": ["title", "nested.object", "nested.machin"],
"typoTolerance": {
"enabled": true,
"disableOnNumbers": true
}
}),
&json!({"q": "12345"}),
|response, code| {
assert_eq!(code, 200, "{}", response);
snapshot!(json_string!(response["hits"]), @r###"
[
{
"id": 2,
"title": "The second document 3398499",
"nested": [
"array",
{
"object": "field 23245121,23223"
},
{
"prout": "truc 123980612321",
"machin": "lol 12345645333447879"
}
]
}
]
"###);
},
)
.await;
// Test typo search
test_settings_documents_indexing_swapping_and_search(
&documents,
&json!({
"searchableAttributes": ["title", "nested.object", "nested.machin"],
"typoTolerance": {
"enabled": true,
"disableOnNumbers": true
}
}),
&json!({"q": "123457"}),
|response, code| {
assert_eq!(code, 200, "{}", response);
snapshot!(json_string!(response["hits"]), @r###"[]"###);
},
)
.await;
}

View File

@@ -274,7 +274,7 @@ async fn settings_bad_typo_tolerance() {
snapshot!(code, @"400 Bad Request");
snapshot!(json_string!(response), @r###"
{
"message": "Unknown field `typoTolerance`: expected one of `enabled`, `minWordSizeForTypos`, `disableOnWords`, `disableOnAttributes`",
"message": "Unknown field `typoTolerance`: expected one of `enabled`, `minWordSizeForTypos`, `disableOnWords`, `disableOnAttributes`, `disableOnNumbers`",
"code": "invalid_settings_typo_tolerance",
"type": "invalid_request",
"link": "https://docs.meilisearch.com/errors#invalid_settings_typo_tolerance"

View File

@@ -179,7 +179,7 @@ test_setting_routes!(
{
setting: typo_tolerance,
update_verb: patch,
default_value: {"enabled": true, "minWordSizeForTypos": {"oneTypo": 5, "twoTypos": 9}, "disableOnWords": [], "disableOnAttributes": []}
default_value: {"enabled": true, "minWordSizeForTypos": {"oneTypo": 5, "twoTypos": 9}, "disableOnWords": [], "disableOnAttributes": [], "disableOnNumbers": false}
},
);
@@ -276,7 +276,7 @@ async fn secrets_are_hidden_in_settings() {
let (response, code) = index.settings().await;
meili_snap::snapshot!(code, @"200 OK");
meili_snap::snapshot!(meili_snap::json_string!(response), @r#"
meili_snap::snapshot!(meili_snap::json_string!(response), @r###"
{
"displayedAttributes": [
"*"
@@ -308,7 +308,8 @@ async fn secrets_are_hidden_in_settings() {
"twoTypos": 9
},
"disableOnWords": [],
"disableOnAttributes": []
"disableOnAttributes": [],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 100,
@@ -337,7 +338,7 @@ async fn secrets_are_hidden_in_settings() {
"facetSearch": true,
"prefixSearch": "indexingTime"
}
"#);
"###);
let (response, code) = server.get_task(settings_update_uid).await;
meili_snap::snapshot!(code, @"200 OK");

View File

@@ -43,7 +43,7 @@ async fn version_too_old() {
std::fs::write(db_path.join("VERSION"), "1.11.9999").unwrap();
let options = Opt { experimental_dumpless_upgrade: true, ..default_settings };
let err = Server::new_with_options(options).await.map(|_| ()).unwrap_err();
snapshot!(err, @"Database version 1.11.9999 is too old for the experimental dumpless upgrade feature. Please generate a dump using the v1.11.9999 and import it in the v1.14.0");
snapshot!(err, @"Database version 1.11.9999 is too old for the experimental dumpless upgrade feature. Please generate a dump using the v1.11.9999 and import it in the v1.15.0");
}
#[actix_rt::test]
@@ -54,11 +54,11 @@ async fn version_requires_downgrade() {
std::fs::create_dir_all(&db_path).unwrap();
let major = meilisearch_types::versioning::VERSION_MAJOR;
let minor = meilisearch_types::versioning::VERSION_MINOR;
let patch = meilisearch_types::versioning::VERSION_PATCH.parse::<u32>().unwrap() + 1;
let patch = meilisearch_types::versioning::VERSION_PATCH + 1;
std::fs::write(db_path.join("VERSION"), format!("{major}.{minor}.{patch}")).unwrap();
let options = Opt { experimental_dumpless_upgrade: true, ..default_settings };
let err = Server::new_with_options(options).await.map(|_| ()).unwrap_err();
snapshot!(err, @"Database version 1.14.1 is higher than the Meilisearch version 1.14.0. Downgrade is not supported");
snapshot!(err, @"Database version 1.15.1 is higher than the Meilisearch version 1.15.0. Downgrade is not supported");
}
#[actix_rt::test]

View File

@@ -1,6 +1,5 @@
---
source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
snapshot_kind: text
---
{
"displayedAttributes": [
@@ -49,7 +48,8 @@ snapshot_kind: text
],
"disableOnAttributes": [
"surname"
]
],
"disableOnNumbers": false
},
"faceting": {
"maxValuesPerFacet": 99,

View File

@@ -8,7 +8,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"progress": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.14.0"
"upgradeTo": "v1.15.0"
},
"stats": {
"totalNbTasks": 1,
@@ -24,7 +24,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task"
"batchCreationComplete": "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task"
},
{
"uid": 23,
@@ -47,7 +47,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.004146631S",
"startedAt": "2025-01-23T11:38:57.012591321Z",
"finishedAt": "2025-01-23T11:38:57.016737952Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 22,
@@ -71,7 +71,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.102738497S",
"startedAt": "2025-01-23T11:36:22.551906856Z",
"finishedAt": "2025-01-23T11:36:22.654645353Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 21,
@@ -95,7 +95,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.005108474S",
"startedAt": "2025-01-23T11:36:04.132670526Z",
"finishedAt": "2025-01-23T11:36:04.137779Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 20,
@@ -119,7 +119,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.027954894S",
"startedAt": "2025-01-23T11:35:53.631082795Z",
"finishedAt": "2025-01-23T11:35:53.659037689Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 19,
@@ -142,7 +142,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.006903297S",
"startedAt": "2025-01-20T11:50:52.874106134Z",
"finishedAt": "2025-01-20T11:50:52.881009431Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 18,
@@ -171,7 +171,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000481257S",
"startedAt": "2025-01-20T11:48:04.92820416Z",
"finishedAt": "2025-01-20T11:48:04.928685417Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 17,
@@ -194,7 +194,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000407005S",
"startedAt": "2025-01-20T11:47:53.509403957Z",
"finishedAt": "2025-01-20T11:47:53.509810962Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 16,
@@ -217,7 +217,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000403716S",
"startedAt": "2025-01-20T11:47:48.430653005Z",
"finishedAt": "2025-01-20T11:47:48.431056721Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 15,
@@ -240,7 +240,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000417016S",
"startedAt": "2025-01-20T11:47:42.429678617Z",
"finishedAt": "2025-01-20T11:47:42.430095633Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 14,
@@ -264,7 +264,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT12.086284842S",
"startedAt": "2025-01-20T11:47:03.092181576Z",
"finishedAt": "2025-01-20T11:47:15.178466418Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 13,
@@ -296,7 +296,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.011506614S",
"startedAt": "2025-01-16T17:18:43.29334923Z",
"finishedAt": "2025-01-16T17:18:43.304855844Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 12,
@@ -324,7 +324,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007640163S",
"startedAt": "2025-01-16T17:02:52.539749853Z",
"finishedAt": "2025-01-16T17:02:52.547390016Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 11,
@@ -347,7 +347,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007307840S",
"startedAt": "2025-01-16T17:01:14.112756687Z",
"finishedAt": "2025-01-16T17:01:14.120064527Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 10,
@@ -375,7 +375,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007391353S",
"startedAt": "2025-01-16T17:00:29.201180268Z",
"finishedAt": "2025-01-16T17:00:29.208571621Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 9,
@@ -403,7 +403,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007445825S",
"startedAt": "2025-01-16T17:00:15.77629445Z",
"finishedAt": "2025-01-16T17:00:15.783740275Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 8,
@@ -436,7 +436,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.012020083S",
"startedAt": "2025-01-16T16:59:42.744086671Z",
"finishedAt": "2025-01-16T16:59:42.756106754Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 7,
@@ -463,7 +463,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007440092S",
"startedAt": "2025-01-16T16:58:41.2155771Z",
"finishedAt": "2025-01-16T16:58:41.223017192Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 6,
@@ -490,7 +490,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007565161S",
"startedAt": "2025-01-16T16:54:51.940332781Z",
"finishedAt": "2025-01-16T16:54:51.947897942Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 5,
@@ -516,7 +516,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.016307263S",
"startedAt": "2025-01-16T16:53:19.913351957Z",
"finishedAt": "2025-01-16T16:53:19.92965922Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
}
],
"total": 23,

View File

@@ -8,7 +8,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"progress": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.14.0"
"upgradeTo": "v1.15.0"
},
"stats": {
"totalNbTasks": 1,
@@ -24,7 +24,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task"
"batchCreationComplete": "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task"
},
{
"uid": 23,
@@ -47,7 +47,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.004146631S",
"startedAt": "2025-01-23T11:38:57.012591321Z",
"finishedAt": "2025-01-23T11:38:57.016737952Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 22,
@@ -71,7 +71,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.102738497S",
"startedAt": "2025-01-23T11:36:22.551906856Z",
"finishedAt": "2025-01-23T11:36:22.654645353Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 21,
@@ -95,7 +95,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.005108474S",
"startedAt": "2025-01-23T11:36:04.132670526Z",
"finishedAt": "2025-01-23T11:36:04.137779Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 20,
@@ -119,7 +119,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.027954894S",
"startedAt": "2025-01-23T11:35:53.631082795Z",
"finishedAt": "2025-01-23T11:35:53.659037689Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 19,
@@ -142,7 +142,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.006903297S",
"startedAt": "2025-01-20T11:50:52.874106134Z",
"finishedAt": "2025-01-20T11:50:52.881009431Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 18,
@@ -171,7 +171,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000481257S",
"startedAt": "2025-01-20T11:48:04.92820416Z",
"finishedAt": "2025-01-20T11:48:04.928685417Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 17,
@@ -194,7 +194,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000407005S",
"startedAt": "2025-01-20T11:47:53.509403957Z",
"finishedAt": "2025-01-20T11:47:53.509810962Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 16,
@@ -217,7 +217,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000403716S",
"startedAt": "2025-01-20T11:47:48.430653005Z",
"finishedAt": "2025-01-20T11:47:48.431056721Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 15,
@@ -240,7 +240,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000417016S",
"startedAt": "2025-01-20T11:47:42.429678617Z",
"finishedAt": "2025-01-20T11:47:42.430095633Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 14,
@@ -264,7 +264,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT12.086284842S",
"startedAt": "2025-01-20T11:47:03.092181576Z",
"finishedAt": "2025-01-20T11:47:15.178466418Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 13,
@@ -296,7 +296,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.011506614S",
"startedAt": "2025-01-16T17:18:43.29334923Z",
"finishedAt": "2025-01-16T17:18:43.304855844Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 12,
@@ -324,7 +324,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007640163S",
"startedAt": "2025-01-16T17:02:52.539749853Z",
"finishedAt": "2025-01-16T17:02:52.547390016Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 11,
@@ -347,7 +347,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007307840S",
"startedAt": "2025-01-16T17:01:14.112756687Z",
"finishedAt": "2025-01-16T17:01:14.120064527Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 10,
@@ -375,7 +375,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007391353S",
"startedAt": "2025-01-16T17:00:29.201180268Z",
"finishedAt": "2025-01-16T17:00:29.208571621Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 9,
@@ -403,7 +403,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007445825S",
"startedAt": "2025-01-16T17:00:15.77629445Z",
"finishedAt": "2025-01-16T17:00:15.783740275Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 8,
@@ -436,7 +436,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.012020083S",
"startedAt": "2025-01-16T16:59:42.744086671Z",
"finishedAt": "2025-01-16T16:59:42.756106754Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 7,
@@ -463,7 +463,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007440092S",
"startedAt": "2025-01-16T16:58:41.2155771Z",
"finishedAt": "2025-01-16T16:58:41.223017192Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 6,
@@ -490,7 +490,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007565161S",
"startedAt": "2025-01-16T16:54:51.940332781Z",
"finishedAt": "2025-01-16T16:54:51.947897942Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 5,
@@ -516,7 +516,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.016307263S",
"startedAt": "2025-01-16T16:53:19.913351957Z",
"finishedAt": "2025-01-16T16:53:19.92965922Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
}
],
"total": 23,

View File

@@ -8,7 +8,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"progress": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.14.0"
"upgradeTo": "v1.15.0"
},
"stats": {
"totalNbTasks": 1,
@@ -24,7 +24,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task"
"batchCreationComplete": "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task"
},
{
"uid": 23,
@@ -47,7 +47,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.004146631S",
"startedAt": "2025-01-23T11:38:57.012591321Z",
"finishedAt": "2025-01-23T11:38:57.016737952Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 22,
@@ -71,7 +71,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.102738497S",
"startedAt": "2025-01-23T11:36:22.551906856Z",
"finishedAt": "2025-01-23T11:36:22.654645353Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 21,
@@ -95,7 +95,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.005108474S",
"startedAt": "2025-01-23T11:36:04.132670526Z",
"finishedAt": "2025-01-23T11:36:04.137779Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 20,
@@ -119,7 +119,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.027954894S",
"startedAt": "2025-01-23T11:35:53.631082795Z",
"finishedAt": "2025-01-23T11:35:53.659037689Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 19,
@@ -142,7 +142,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.006903297S",
"startedAt": "2025-01-20T11:50:52.874106134Z",
"finishedAt": "2025-01-20T11:50:52.881009431Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 18,
@@ -171,7 +171,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000481257S",
"startedAt": "2025-01-20T11:48:04.92820416Z",
"finishedAt": "2025-01-20T11:48:04.928685417Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 17,
@@ -194,7 +194,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000407005S",
"startedAt": "2025-01-20T11:47:53.509403957Z",
"finishedAt": "2025-01-20T11:47:53.509810962Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 16,
@@ -217,7 +217,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000403716S",
"startedAt": "2025-01-20T11:47:48.430653005Z",
"finishedAt": "2025-01-20T11:47:48.431056721Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 15,
@@ -240,7 +240,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000417016S",
"startedAt": "2025-01-20T11:47:42.429678617Z",
"finishedAt": "2025-01-20T11:47:42.430095633Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 14,
@@ -264,7 +264,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT12.086284842S",
"startedAt": "2025-01-20T11:47:03.092181576Z",
"finishedAt": "2025-01-20T11:47:15.178466418Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 13,
@@ -296,7 +296,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.011506614S",
"startedAt": "2025-01-16T17:18:43.29334923Z",
"finishedAt": "2025-01-16T17:18:43.304855844Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 12,
@@ -324,7 +324,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007640163S",
"startedAt": "2025-01-16T17:02:52.539749853Z",
"finishedAt": "2025-01-16T17:02:52.547390016Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 11,
@@ -347,7 +347,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007307840S",
"startedAt": "2025-01-16T17:01:14.112756687Z",
"finishedAt": "2025-01-16T17:01:14.120064527Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 10,
@@ -375,7 +375,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007391353S",
"startedAt": "2025-01-16T17:00:29.201180268Z",
"finishedAt": "2025-01-16T17:00:29.208571621Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 9,
@@ -403,7 +403,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007445825S",
"startedAt": "2025-01-16T17:00:15.77629445Z",
"finishedAt": "2025-01-16T17:00:15.783740275Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 8,
@@ -436,7 +436,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.012020083S",
"startedAt": "2025-01-16T16:59:42.744086671Z",
"finishedAt": "2025-01-16T16:59:42.756106754Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 7,
@@ -463,7 +463,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007440092S",
"startedAt": "2025-01-16T16:58:41.2155771Z",
"finishedAt": "2025-01-16T16:58:41.223017192Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 6,
@@ -490,7 +490,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007565161S",
"startedAt": "2025-01-16T16:54:51.940332781Z",
"finishedAt": "2025-01-16T16:54:51.947897942Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 5,
@@ -516,7 +516,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.016307263S",
"startedAt": "2025-01-16T16:53:19.913351957Z",
"finishedAt": "2025-01-16T16:53:19.92965922Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
}
],
"total": 23,

View File

@@ -29,7 +29,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
}
],
"total": 1,

View File

@@ -25,7 +25,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 0,
@@ -49,7 +49,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.111055654S",
"startedAt": "2025-01-16T16:45:16.020248085Z",
"finishedAt": "2025-01-16T16:45:16.131303739Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
}
],
"total": 2,

View File

@@ -25,7 +25,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 0,
@@ -49,7 +49,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.111055654S",
"startedAt": "2025-01-16T16:45:16.020248085Z",
"finishedAt": "2025-01-16T16:45:16.131303739Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
}
],
"total": 2,

View File

@@ -25,7 +25,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 0,
@@ -49,7 +49,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.111055654S",
"startedAt": "2025-01-16T16:45:16.020248085Z",
"finishedAt": "2025-01-16T16:45:16.131303739Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
}
],
"total": 2,

View File

@@ -30,7 +30,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
}
],
"total": 1,

View File

@@ -30,7 +30,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
}
],
"total": 1,

View File

@@ -29,7 +29,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
}
],
"total": 1,

View File

@@ -12,7 +12,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"canceledBy": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.14.0"
"upgradeTo": "v1.15.0"
},
"error": null,
"duration": "[duration]",

View File

@@ -12,7 +12,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"canceledBy": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.14.0"
"upgradeTo": "v1.15.0"
},
"error": null,
"duration": "[duration]",

View File

@@ -12,7 +12,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"canceledBy": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.14.0"
"upgradeTo": "v1.15.0"
},
"error": null,
"duration": "[duration]",

View File

@@ -8,7 +8,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"progress": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.14.0"
"upgradeTo": "v1.15.0"
},
"stats": {
"totalNbTasks": 1,
@@ -24,7 +24,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "[duration]",
"startedAt": "[date]",
"finishedAt": "[date]",
"batcherStoppedBecause": "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task"
"batchCreationComplete": "a batch of tasks of type `upgradeDatabase` cannot be batched with any other type of task"
},
{
"uid": 23,
@@ -47,7 +47,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.004146631S",
"startedAt": "2025-01-23T11:38:57.012591321Z",
"finishedAt": "2025-01-23T11:38:57.016737952Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 22,
@@ -71,7 +71,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.102738497S",
"startedAt": "2025-01-23T11:36:22.551906856Z",
"finishedAt": "2025-01-23T11:36:22.654645353Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 21,
@@ -95,7 +95,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.005108474S",
"startedAt": "2025-01-23T11:36:04.132670526Z",
"finishedAt": "2025-01-23T11:36:04.137779Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 20,
@@ -119,7 +119,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.027954894S",
"startedAt": "2025-01-23T11:35:53.631082795Z",
"finishedAt": "2025-01-23T11:35:53.659037689Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 19,
@@ -142,7 +142,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.006903297S",
"startedAt": "2025-01-20T11:50:52.874106134Z",
"finishedAt": "2025-01-20T11:50:52.881009431Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 18,
@@ -171,7 +171,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000481257S",
"startedAt": "2025-01-20T11:48:04.92820416Z",
"finishedAt": "2025-01-20T11:48:04.928685417Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 17,
@@ -194,7 +194,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000407005S",
"startedAt": "2025-01-20T11:47:53.509403957Z",
"finishedAt": "2025-01-20T11:47:53.509810962Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 16,
@@ -217,7 +217,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000403716S",
"startedAt": "2025-01-20T11:47:48.430653005Z",
"finishedAt": "2025-01-20T11:47:48.431056721Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 15,
@@ -240,7 +240,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.000417016S",
"startedAt": "2025-01-20T11:47:42.429678617Z",
"finishedAt": "2025-01-20T11:47:42.430095633Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 14,
@@ -264,7 +264,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT12.086284842S",
"startedAt": "2025-01-20T11:47:03.092181576Z",
"finishedAt": "2025-01-20T11:47:15.178466418Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 13,
@@ -296,7 +296,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.011506614S",
"startedAt": "2025-01-16T17:18:43.29334923Z",
"finishedAt": "2025-01-16T17:18:43.304855844Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 12,
@@ -324,7 +324,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007640163S",
"startedAt": "2025-01-16T17:02:52.539749853Z",
"finishedAt": "2025-01-16T17:02:52.547390016Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 11,
@@ -347,7 +347,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007307840S",
"startedAt": "2025-01-16T17:01:14.112756687Z",
"finishedAt": "2025-01-16T17:01:14.120064527Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 10,
@@ -375,7 +375,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007391353S",
"startedAt": "2025-01-16T17:00:29.201180268Z",
"finishedAt": "2025-01-16T17:00:29.208571621Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 9,
@@ -403,7 +403,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007445825S",
"startedAt": "2025-01-16T17:00:15.77629445Z",
"finishedAt": "2025-01-16T17:00:15.783740275Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 8,
@@ -436,7 +436,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.012020083S",
"startedAt": "2025-01-16T16:59:42.744086671Z",
"finishedAt": "2025-01-16T16:59:42.756106754Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 7,
@@ -463,7 +463,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007440092S",
"startedAt": "2025-01-16T16:58:41.2155771Z",
"finishedAt": "2025-01-16T16:58:41.223017192Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 6,
@@ -490,7 +490,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007565161S",
"startedAt": "2025-01-16T16:54:51.940332781Z",
"finishedAt": "2025-01-16T16:54:51.947897942Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 5,
@@ -516,7 +516,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.016307263S",
"startedAt": "2025-01-16T16:53:19.913351957Z",
"finishedAt": "2025-01-16T16:53:19.92965922Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 4,
@@ -540,7 +540,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.087655941S",
"startedAt": "2025-01-16T16:52:32.631145531Z",
"finishedAt": "2025-01-16T16:52:32.718801472Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 3,
@@ -565,7 +565,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.007593573S",
"startedAt": "2025-01-16T16:47:53.677901409Z",
"finishedAt": "2025-01-16T16:47:53.685494982Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 2,
@@ -591,7 +591,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.017769760S",
"startedAt": "2025-01-16T16:47:41.211587682Z",
"finishedAt": "2025-01-16T16:47:41.229357442Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 1,
@@ -615,7 +615,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.066095506S",
"startedAt": "2025-01-16T16:47:10.217299609Z",
"finishedAt": "2025-01-16T16:47:10.283395115Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
},
{
"uid": 0,
@@ -639,7 +639,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"duration": "PT0.111055654S",
"startedAt": "2025-01-16T16:45:16.020248085Z",
"finishedAt": "2025-01-16T16:45:16.131303739Z",
"batcherStoppedBecause": "unspecified"
"batchCreationComplete": "unspecified"
}
],
"total": 25,

View File

@@ -12,7 +12,7 @@ source: crates/meilisearch/tests/upgrade/v1_12/v1_12_0.rs
"canceledBy": null,
"details": {
"upgradeFrom": "v1.12.0",
"upgradeTo": "v1.14.0"
"upgradeTo": "v1.15.0"
},
"error": null,
"duration": "[duration]",

View File

@@ -49,15 +49,10 @@ impl OfflineUpgrade {
const LAST_SUPPORTED_UPGRADE_TO_VERSION: &str = "1.12.7";
let upgrade_list = [
(
v1_9_to_v1_10 as fn(&Path, u32, u32, u32) -> Result<(), anyhow::Error>,
"1",
"10",
"0",
),
(v1_10_to_v1_11, "1", "11", "0"),
(v1_11_to_v1_12, "1", "12", "0"),
(v1_12_to_v1_12_3, "1", "12", "3"),
(v1_9_to_v1_10 as fn(&Path, u32, u32, u32) -> Result<(), anyhow::Error>, 1, 10, 0),
(v1_10_to_v1_11, 1, 11, 0),
(v1_11_to_v1_12, 1, 12, 0),
(v1_12_to_v1_12_3, 1, 12, 3),
];
let no_upgrade: usize = upgrade_list.len();
@@ -95,13 +90,8 @@ impl OfflineUpgrade {
if start_at == no_upgrade {
println!("No upgrade operation to perform, writing VERSION file");
create_version_file(
&self.db_path,
&target_major.to_string(),
&target_minor.to_string(),
&target_patch.to_string(),
)
.context("while writing VERSION file after the upgrade")?;
create_version_file(&self.db_path, target_major, target_minor, target_patch)
.context("while writing VERSION file after the upgrade")?;
println!("Success");
return Ok(());
}

View File

@@ -1,6 +1,13 @@
pub static VERSION_MAJOR: &str = env!("CARGO_PKG_VERSION_MAJOR");
pub static VERSION_MINOR: &str = env!("CARGO_PKG_VERSION_MINOR");
pub static VERSION_PATCH: &str = env!("CARGO_PKG_VERSION_PATCH");
pub const VERSION_MAJOR: u32 = parse_u32(env!("CARGO_PKG_VERSION_MAJOR"));
pub const VERSION_MINOR: u32 = parse_u32(env!("CARGO_PKG_VERSION_MINOR"));
pub const VERSION_PATCH: u32 = parse_u32(env!("CARGO_PKG_VERSION_PATCH"));
const fn parse_u32(s: &str) -> u32 {
match u32::from_str_radix(s, 10) {
Ok(version) => version,
Err(_) => panic!("could not parse as u32"),
}
}
pub const RESERVED_VECTORS_FIELD_NAME: &str = "_vectors";
pub const RESERVED_GEO_FIELD_NAME: &str = "_geo";

View File

@@ -0,0 +1,50 @@
use heed::{
types::{SerdeJson, Str},
RoTxn, RwTxn,
};
use serde::{Deserialize, Serialize};
use crate::{index::main_key, Index};
#[derive(Debug, Clone, Copy, Serialize, Deserialize, PartialEq, Eq, Default)]
#[serde(rename_all = "camelCase")]
pub struct DisabledTyposTerms {
pub disable_on_numbers: bool,
}
impl Index {
pub fn disabled_typos_terms(&self, txn: &RoTxn<'_>) -> heed::Result<DisabledTyposTerms> {
self.main
.remap_types::<Str, SerdeJson<DisabledTyposTerms>>()
.get(txn, main_key::DISABLED_TYPOS_TERMS)
.map(|option| option.unwrap_or_default())
}
pub(crate) fn put_disabled_typos_terms(
&self,
txn: &mut RwTxn<'_>,
disabled_typos_terms: &DisabledTyposTerms,
) -> heed::Result<()> {
self.main.remap_types::<Str, SerdeJson<DisabledTyposTerms>>().put(
txn,
main_key::DISABLED_TYPOS_TERMS,
disabled_typos_terms,
)?;
Ok(())
}
pub(crate) fn delete_disabled_typos_terms(&self, txn: &mut RwTxn<'_>) -> heed::Result<()> {
self.main
.remap_types::<Str, SerdeJson<DisabledTyposTerms>>()
.delete(txn, main_key::DISABLED_TYPOS_TERMS)?;
Ok(())
}
}
impl DisabledTyposTerms {
pub fn is_exact(&self, word: &str) -> bool {
// If disable_on_numbers is true, we disable the word if it contains only numbers or punctuation
self.disable_on_numbers && word.chars().all(|c| c.is_numeric() || c.is_ascii_punctuation())
}
}

View File

@@ -32,13 +32,13 @@ impl ExternalDocumentsIds {
&self,
rtxn: &RoTxn<'_>,
external_id: A,
) -> heed::Result<Option<u32>> {
) -> heed::Result<Option<DocumentId>> {
self.0.get(rtxn, external_id.as_ref())
}
/// An helper function to debug this type, returns an `HashMap` of both,
/// soft and hard fst maps, combined.
pub fn to_hash_map(&self, rtxn: &RoTxn<'_>) -> heed::Result<HashMap<String, u32>> {
pub fn to_hash_map(&self, rtxn: &RoTxn<'_>) -> heed::Result<HashMap<String, DocumentId>> {
let mut map = HashMap::default();
for result in self.0.iter(rtxn)? {
let (external, internal) = result?;

View File

@@ -7,6 +7,7 @@ use crate::FieldId;
mod global;
pub mod metadata;
pub use global::GlobalFieldsIdsMap;
pub use metadata::{FieldIdMapWithMetadata, MetadataBuilder};
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct FieldsIdsMap {

File diff suppressed because it is too large Load Diff

View File

@@ -12,6 +12,7 @@ mod asc_desc;
mod attribute_patterns;
mod criterion;
pub mod database_stats;
pub mod disabled_typos_terms;
mod error;
mod external_documents_ids;
pub mod facet;
@@ -51,18 +52,19 @@ pub use search::new::{
};
use serde_json::Value;
pub use thread_pool_no_abort::{PanicCatched, ThreadPoolNoAbort, ThreadPoolNoAbortBuilder};
pub use {charabia as tokenizer, heed, rhai};
pub use {arroy, charabia as tokenizer, heed, rhai};
pub use self::asc_desc::{AscDesc, AscDescError, Member, SortError};
pub use self::attribute_patterns::AttributePatterns;
pub use self::attribute_patterns::PatternMatch;
pub use self::attribute_patterns::{AttributePatterns, PatternMatch};
pub use self::criterion::{default_criteria, Criterion, CriterionError};
pub use self::error::{
Error, FieldIdMapMissingEntry, InternalError, SerializationError, UserError,
};
pub use self::external_documents_ids::ExternalDocumentsIds;
pub use self::fieldids_weights_map::FieldidsWeightsMap;
pub use self::fields_ids_map::{FieldsIdsMap, GlobalFieldsIdsMap};
pub use self::fields_ids_map::{
FieldIdMapWithMetadata, FieldsIdsMap, GlobalFieldsIdsMap, MetadataBuilder,
};
pub use self::filterable_attributes_rules::{
FilterFeatures, FilterableAttributesFeatures, FilterableAttributesPatterns,
FilterableAttributesRule,
@@ -83,8 +85,6 @@ pub use self::search::{
};
pub use self::update::ChannelCongestion;
pub use arroy;
pub type Result<T> = std::result::Result<T, error::Error>;
pub type Attribute = u32;

View File

@@ -105,10 +105,10 @@ impl Prompt {
max_bytes,
};
// render template with special object that's OK with `doc.*` and `fields.*`
this.template
.render(&template_checker::TemplateChecker)
.map_err(NewPromptError::invalid_fields_in_template)?;
// // render template with special object that's OK with `doc.*` and `fields.*`
// this.template
// .render(&template_checker::TemplateChecker)
// .map_err(NewPromptError::invalid_fields_in_template)?;
Ok(this)
}

View File

@@ -1,10 +1,11 @@
use std::collections::BTreeSet;
use std::fmt::{Debug, Display};
use std::ops::Bound::{self, Excluded, Included};
use std::ops::Bound::{self, Excluded, Included, Unbounded};
use either::Either;
pub use filter_parser::{Condition, Error as FPError, FilterCondition, Token};
use heed::types::LazyDecode;
use heed::BytesEncode;
use memchr::memmem::Finder;
use roaring::{MultiOps, RoaringBitmap};
use serde_json::Value;
@@ -14,7 +15,7 @@ use crate::constants::RESERVED_GEO_FIELD_NAME;
use crate::error::{Error, UserError};
use crate::filterable_attributes_rules::{filtered_matching_patterns, matching_features};
use crate::heed_codec::facet::{
FacetGroupKey, FacetGroupKeyCodec, FacetGroupValue, FacetGroupValueCodec, OrderedF64Codec,
FacetGroupKey, FacetGroupKeyCodec, FacetGroupValue, FacetGroupValueCodec,
};
use crate::index::db_name::FACET_ID_STRING_DOCIDS;
use crate::{
@@ -271,7 +272,7 @@ impl<'a> Filter<'a> {
// as the facets values are all in the same database and prefixed by the
// field id and the level.
let (left, right) = match operator {
let (number_bounds, (left_str, right_str)) = match operator {
// return an error if the filter is not allowed for this field
Condition::GreaterThan(_)
| Condition::GreaterThanOrEqual(_)
@@ -305,17 +306,37 @@ impl<'a> Filter<'a> {
));
}
Condition::GreaterThan(val) => {
(Excluded(val.parse_finite_float()?), Included(f64::MAX))
let number = val.parse_finite_float().ok();
let number_bounds = number.map(|number| (Excluded(number), Included(f64::MAX)));
let str_bounds = (Excluded(val.value()), Unbounded);
(number_bounds, str_bounds)
}
Condition::GreaterThanOrEqual(val) => {
(Included(val.parse_finite_float()?), Included(f64::MAX))
let number = val.parse_finite_float().ok();
let number_bounds = number.map(|number| (Included(number), Included(f64::MAX)));
let str_bounds = (Included(val.value()), Unbounded);
(number_bounds, str_bounds)
}
Condition::LowerThan(val) => {
let number = val.parse_finite_float().ok();
let number_bounds = number.map(|number| (Included(f64::MIN), Excluded(number)));
let str_bounds = (Unbounded, Excluded(val.value()));
(number_bounds, str_bounds)
}
Condition::LowerThan(val) => (Included(f64::MIN), Excluded(val.parse_finite_float()?)),
Condition::LowerThanOrEqual(val) => {
(Included(f64::MIN), Included(val.parse_finite_float()?))
let number = val.parse_finite_float().ok();
let number_bounds = number.map(|number| (Included(f64::MIN), Included(number)));
let str_bounds = (Unbounded, Included(val.value()));
(number_bounds, str_bounds)
}
Condition::Between { from, to } => {
(Included(from.parse_finite_float()?), Included(to.parse_finite_float()?))
let from_number = from.parse_finite_float().ok();
let to_number = to.parse_finite_float().ok();
let number_bounds =
from_number.zip(to_number).map(|(from, to)| (Included(from), Included(to)));
let str_bounds = (Included(from.value()), Included(to.value()));
(number_bounds, str_bounds)
}
Condition::Null => {
let is_null = index.null_faceted_documents_ids(rtxn, field_id)?;
@@ -415,29 +436,47 @@ impl<'a> Filter<'a> {
};
let mut output = RoaringBitmap::new();
Self::explore_facet_number_levels(
if let Some((left_number, right_number)) = number_bounds {
Self::explore_facet_levels(
rtxn,
numbers_db,
field_id,
&left_number,
&right_number,
universe,
&mut output,
)?;
}
Self::explore_facet_levels(
rtxn,
numbers_db,
strings_db,
field_id,
left,
right,
&left_str,
&right_str,
universe,
&mut output,
)?;
Ok(output)
}
/// Aggregates the documents ids that are part of the specified range automatically
/// going deeper through the levels.
fn explore_facet_number_levels(
rtxn: &heed::RoTxn<'_>,
db: heed::Database<FacetGroupKeyCodec<OrderedF64Codec>, FacetGroupValueCodec>,
fn explore_facet_levels<'data, BoundCodec>(
rtxn: &'data heed::RoTxn<'data>,
db: heed::Database<FacetGroupKeyCodec<BoundCodec>, FacetGroupValueCodec>,
field_id: FieldId,
left: Bound<f64>,
right: Bound<f64>,
left: &'data Bound<<BoundCodec as heed::BytesEncode<'data>>::EItem>,
right: &'data Bound<<BoundCodec as heed::BytesEncode<'data>>::EItem>,
universe: Option<&RoaringBitmap>,
output: &mut RoaringBitmap,
) -> Result<()> {
) -> Result<()>
where
BoundCodec: for<'b> BytesEncode<'b>,
for<'b> <BoundCodec as BytesEncode<'b>>::EItem: Sized + PartialOrd,
{
match (left, right) {
// lower TO upper when lower > upper must return no result
(Included(l), Included(r)) if l > r => return Ok(()),
@@ -446,8 +485,8 @@ impl<'a> Filter<'a> {
(Excluded(l), Included(r)) if l >= r => return Ok(()),
(_, _) => (),
}
facet_range_search::find_docids_of_facet_within_bounds::<OrderedF64Codec>(
rtxn, db, field_id, &left, &right, universe, output,
facet_range_search::find_docids_of_facet_within_bounds::<BoundCodec>(
rtxn, db, field_id, left, right, universe, output,
)?;
Ok(())
@@ -1249,28 +1288,24 @@ mod tests {
let result = filter.evaluate(&rtxn, &index).unwrap();
assert!(result.contains(0));
let filter = Filter::from_str("price < inf").unwrap().unwrap();
assert!(matches!(
filter.evaluate(&rtxn, &index),
Err(crate::Error::UserError(crate::error::UserError::InvalidFilter(_)))
));
let result = filter.evaluate(&rtxn, &index).unwrap();
// this is allowed due to filters with strings
assert!(result.contains(1));
let filter = Filter::from_str("price = NaN").unwrap().unwrap();
let result = filter.evaluate(&rtxn, &index).unwrap();
assert!(result.is_empty());
let filter = Filter::from_str("price < NaN").unwrap().unwrap();
assert!(matches!(
filter.evaluate(&rtxn, &index),
Err(crate::Error::UserError(crate::error::UserError::InvalidFilter(_)))
));
let result = filter.evaluate(&rtxn, &index).unwrap();
assert!(result.contains(1));
let filter = Filter::from_str("price = infinity").unwrap().unwrap();
let result = filter.evaluate(&rtxn, &index).unwrap();
assert!(result.contains(2));
let filter = Filter::from_str("price < infinity").unwrap().unwrap();
assert!(matches!(
filter.evaluate(&rtxn, &index),
Err(crate::Error::UserError(crate::error::UserError::InvalidFilter(_)))
));
let result = filter.evaluate(&rtxn, &index).unwrap();
assert!(result.contains(0));
assert!(result.contains(1));
}
#[test]

View File

@@ -8,7 +8,7 @@ use roaring::bitmap::RoaringBitmap;
pub use self::facet::{FacetDistribution, Filter, OrderBy, DEFAULT_VALUES_PER_FACET};
pub use self::new::matches::{FormatOptions, MatchBounds, MatcherBuilder, MatchingWords};
use self::new::{execute_vector_search, PartialSearchResult};
use self::new::{execute_vector_search, PartialSearchResult, VectorStoreStats};
use crate::filterable_attributes_rules::{filtered_matching_patterns, matching_features};
use crate::score_details::{ScoreDetails, ScoringStrategy};
use crate::vector::Embedder;
@@ -269,6 +269,12 @@ impl<'a> Search<'a> {
)?,
};
if let Some(VectorStoreStats { total_time, total_queries, total_results }) =
ctx.vector_store_stats
{
tracing::debug!("Vector store stats: total_time={total_time:.02?}, total_queries={total_queries}, total_results={total_results}");
}
// consume context and located_query_terms to build MatchingWords.
let matching_words = match located_query_terms {
Some(located_query_terms) => MatchingWords::new(ctx, located_query_terms),

View File

@@ -22,6 +22,8 @@ mod vector_sort;
mod tests;
use std::collections::HashSet;
use std::ops::AddAssign;
use std::time::Duration;
use bucket_sort::{bucket_sort, BucketSortOutput};
use charabia::{Language, TokenizerBuilder};
@@ -72,6 +74,7 @@ pub struct SearchContext<'ctx> {
pub phrase_docids: PhraseDocIdsCache,
pub restricted_fids: Option<RestrictedFids>,
pub prefix_search: PrefixSearch,
pub vector_store_stats: Option<VectorStoreStats>,
}
impl<'ctx> SearchContext<'ctx> {
@@ -101,6 +104,7 @@ impl<'ctx> SearchContext<'ctx> {
phrase_docids: <_>::default(),
restricted_fids: None,
prefix_search,
vector_store_stats: None,
})
}
@@ -166,6 +170,25 @@ impl<'ctx> SearchContext<'ctx> {
}
}
#[derive(Debug, Default)]
pub struct VectorStoreStats {
/// The total time spent on vector search.
pub total_time: Duration,
/// The number of searches performed.
pub total_queries: usize,
/// The number of nearest neighbors found.
pub total_results: usize,
}
impl AddAssign for VectorStoreStats {
fn add_assign(&mut self, other: Self) {
let Self { total_time, total_queries, total_results } = self;
*total_time += other.total_time;
*total_queries += other.total_queries;
*total_results += other.total_results;
}
}
#[derive(Clone, Copy, PartialEq, PartialOrd, Ord, Eq)]
pub enum Word {
Original(Interned<String>),

View File

@@ -1,8 +1,10 @@
use std::iter::FromIterator;
use std::time::Instant;
use roaring::RoaringBitmap;
use super::ranking_rules::{RankingRule, RankingRuleOutput, RankingRuleQueryTrait};
use super::VectorStoreStats;
use crate::score_details::{self, ScoreDetails};
use crate::vector::{ArroyWrapper, DistributionShift, Embedder};
use crate::{DocumentId, Result, SearchContext, SearchLogger};
@@ -53,9 +55,15 @@ impl<Q: RankingRuleQueryTrait> VectorSort<Q> {
) -> Result<()> {
let target = &self.target;
let before = Instant::now();
let reader = ArroyWrapper::new(ctx.index.vector_arroy, self.embedder_index, self.quantized);
let results = reader.nns_by_vector(ctx.txn, target, self.limit, Some(vector_candidates))?;
self.cached_sorted_docids = results.into_iter();
*ctx.vector_store_stats.get_or_insert_default() += VectorStoreStats {
total_time: before.elapsed(),
total_queries: 1,
total_results: self.cached_sorted_docids.len(),
};
Ok(())
}

View File

@@ -1,4 +0,0 @@
---
source: milli/src/index.rs
---
[0, ]

View File

@@ -1,4 +0,0 @@
---
source: milli/src/index.rs
---
[]

View File

@@ -0,0 +1,4 @@
---
source: crates/milli/src/test_index.rs
---
[0, ]

View File

@@ -0,0 +1,4 @@
---
source: crates/milli/src/test_index.rs
---
[]

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,45 @@
use deserr::Deserr;
use serde::{Deserialize, Serialize};
use utoipa::ToSchema;
use crate::index::ChatConfig;
use crate::prompt::{default_max_bytes, PromptData};
use crate::update::Setting;
#[derive(Debug, Clone, Default, Serialize, Deserialize, PartialEq, Eq, Deserr, ToSchema)]
#[serde(deny_unknown_fields, rename_all = "camelCase")]
#[deserr(deny_unknown_fields, rename_all = camelCase)]
pub struct ChatSettings {
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<String>)]
pub description: Setting<String>,
/// A liquid template used to render documents to a text that can be embedded.
///
/// Meillisearch interpolates the template for each document and sends the resulting text to the embedder.
/// The embedder then generates document vectors based on this text.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<String>)]
pub document_template: Setting<String>,
/// Rendered texts are truncated to this size. Defaults to 400.
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<usize>)]
pub document_template_max_bytes: Setting<usize>,
}
impl From<ChatConfig> for ChatSettings {
fn from(config: ChatConfig) -> Self {
let ChatConfig { description, prompt: PromptData { template, max_bytes } } = config;
ChatSettings {
description: Setting::Set(description),
document_template: Setting::Set(template),
document_template_max_bytes: Setting::Set(
max_bytes.unwrap_or(default_max_bytes()).get(),
),
}
}
}

View File

@@ -127,7 +127,8 @@ pub fn extract_word_docids<R: io::Read + io::Seek>(
// merge all deletions
let obkv = KvReaderDelAdd::from_slice(value);
if let Some(value) = obkv.get(DelAdd::Deletion) {
let delete_from_exact = settings_diff.old.exact_attributes.contains(&fid);
let delete_from_exact = settings_diff.old.exact_attributes.contains(&fid)
|| settings_diff.old.disabled_typos_terms.is_exact(w);
buffer.clear();
let mut obkv = KvWriterDelAdd::new(&mut buffer);
obkv.insert(DelAdd::Deletion, value)?;
@@ -139,7 +140,8 @@ pub fn extract_word_docids<R: io::Read + io::Seek>(
}
// merge all additions
if let Some(value) = obkv.get(DelAdd::Addition) {
let add_in_exact = settings_diff.new.exact_attributes.contains(&fid);
let add_in_exact = settings_diff.new.exact_attributes.contains(&fid)
|| settings_diff.new.disabled_typos_terms.is_exact(w);
buffer.clear();
let mut obkv = KvWriterDelAdd::new(&mut buffer);
obkv.insert(DelAdd::Addition, value)?;

View File

@@ -273,14 +273,11 @@ pub(crate) fn write_typed_chunk_into_index(
unreachable!();
};
let clonable_word_docids = unsafe { as_cloneable_grenad(&word_docids_reader) }?;
let clonable_exact_word_docids =
unsafe { as_cloneable_grenad(&exact_word_docids_reader) }?;
word_docids_builder.push(word_docids_reader.into_cursor()?);
exact_word_docids_builder.push(exact_word_docids_reader.into_cursor()?);
word_fid_docids_builder.push(word_fid_docids_reader.into_cursor()?);
fst_merger_builder.push(clonable_word_docids.into_cursor()?);
fst_merger_builder.push(clonable_exact_word_docids.into_cursor()?);
}
let word_docids_merger = word_docids_builder.build();

View File

@@ -1,4 +1,5 @@
pub use self::available_ids::AvailableIds;
pub use self::chat::ChatSettings;
pub use self::clear_documents::ClearDocuments;
pub use self::concurrent_available_ids::ConcurrentAvailableIds;
pub use self::facet::bulk::FacetsUpdateBulk;
@@ -13,6 +14,7 @@ pub use self::words_prefix_integer_docids::WordPrefixIntegerDocids;
pub use self::words_prefixes_fst::WordsPrefixesFst;
mod available_ids;
mod chat;
mod clear_documents;
mod concurrent_available_ids;
pub(crate) mod del_add;

View File

@@ -319,8 +319,11 @@ impl WordDocidsExtractors {
let doc_alloc = &context.doc_alloc;
let exact_attributes = index.exact_attributes(rtxn)?;
let is_exact_attribute =
|fname: &str| exact_attributes.iter().any(|attr| contained_in(fname, attr));
let disabled_typos_terms = index.disabled_typos_terms(rtxn)?;
let is_exact = |fname: &str, word: &str| {
exact_attributes.iter().any(|attr| contained_in(fname, attr))
|| disabled_typos_terms.is_exact(word)
};
match document_change {
DocumentChange::Deletion(inner) => {
let mut token_fn = |fname: &str, fid, pos, word: &str| {
@@ -328,7 +331,7 @@ impl WordDocidsExtractors {
fid,
pos,
word,
is_exact_attribute(fname),
is_exact(fname, word),
inner.docid(),
doc_alloc,
)
@@ -356,7 +359,7 @@ impl WordDocidsExtractors {
fid,
pos,
word,
is_exact_attribute(fname),
is_exact(fname, word),
inner.docid(),
doc_alloc,
)
@@ -372,7 +375,7 @@ impl WordDocidsExtractors {
fid,
pos,
word,
is_exact_attribute(fname),
is_exact(fname, word),
inner.docid(),
doc_alloc,
)
@@ -389,7 +392,7 @@ impl WordDocidsExtractors {
fid,
pos,
word,
is_exact_attribute(fname),
is_exact(fname, word),
inner.docid(),
doc_alloc,
)

View File

@@ -9,6 +9,7 @@ pub use document_operation::{DocumentOperation, PayloadStats};
use hashbrown::HashMap;
use heed::RwTxn;
pub use partial_dump::PartialDump;
pub use post_processing::recompute_word_fst_from_word_docids_database;
pub use update_by_function::UpdateByFunction;
pub use write::ChannelCongestion;
use write::{build_vectors, update_index, write_to_db};

View File

@@ -131,6 +131,20 @@ fn compute_word_fst(
}
}
pub fn recompute_word_fst_from_word_docids_database(index: &Index, wtxn: &mut RwTxn) -> Result<()> {
let fst = fst::Set::default().map_data(std::borrow::Cow::Owned)?;
let mut word_fst_builder = WordFstBuilder::new(&fst)?;
let words = index.word_docids.iter(wtxn)?.remap_data_type::<DecodeIgnore>();
for res in words {
let (word, _) = res?;
word_fst_builder.register_word(DelAdd::Addition, word.as_ref())?;
}
let (word_fst_mmap, _) = word_fst_builder.build(index, wtxn)?;
index.main.remap_types::<Str, Bytes>().put(wtxn, WORDS_FST_KEY, &word_fst_mmap)?;
Ok(())
}
#[tracing::instrument(level = "trace", skip_all, target = "indexing::facet_search")]
fn compute_facet_search_database(
index: &Index,

View File

@@ -13,19 +13,20 @@ use time::OffsetDateTime;
use super::del_add::{DelAdd, DelAddOperation};
use super::index_documents::{IndexDocumentsConfig, Transform};
use super::IndexerConfig;
use super::{ChatSettings, IndexerConfig};
use crate::attribute_patterns::PatternMatch;
use crate::constants::RESERVED_GEO_FIELD_NAME;
use crate::criterion::Criterion;
use crate::disabled_typos_terms::DisabledTyposTerms;
use crate::error::UserError;
use crate::fields_ids_map::metadata::{FieldIdMapWithMetadata, MetadataBuilder};
use crate::filterable_attributes_rules::match_faceted_field;
use crate::index::{
IndexEmbeddingConfig, PrefixSearch, DEFAULT_MIN_WORD_LEN_ONE_TYPO,
ChatConfig, IndexEmbeddingConfig, PrefixSearch, DEFAULT_MIN_WORD_LEN_ONE_TYPO,
DEFAULT_MIN_WORD_LEN_TWO_TYPOS,
};
use crate::order_by_map::OrderByMap;
use crate::prompt::default_max_bytes;
use crate::prompt::{default_max_bytes, PromptData};
use crate::proximity::ProximityPrecision;
use crate::update::index_documents::IndexDocumentsMethod;
use crate::update::{IndexDocuments, UpdateIndexingStep};
@@ -169,6 +170,7 @@ pub struct Settings<'a, 't, 'i> {
synonyms: Setting<BTreeMap<String, Vec<String>>>,
primary_key: Setting<String>,
authorize_typos: Setting<bool>,
disable_on_numbers: Setting<bool>,
min_word_len_two_typos: Setting<u8>,
min_word_len_one_typo: Setting<u8>,
exact_words: Setting<BTreeSet<String>>,
@@ -183,6 +185,7 @@ pub struct Settings<'a, 't, 'i> {
localized_attributes_rules: Setting<Vec<LocalizedAttributesRule>>,
prefix_search: Setting<PrefixSearch>,
facet_search: Setting<bool>,
chat: Setting<ChatSettings>,
}
impl<'a, 't, 'i> Settings<'a, 't, 'i> {
@@ -207,6 +210,7 @@ impl<'a, 't, 'i> Settings<'a, 't, 'i> {
synonyms: Setting::NotSet,
primary_key: Setting::NotSet,
authorize_typos: Setting::NotSet,
disable_on_numbers: Setting::NotSet,
exact_words: Setting::NotSet,
min_word_len_two_typos: Setting::NotSet,
min_word_len_one_typo: Setting::NotSet,
@@ -220,6 +224,7 @@ impl<'a, 't, 'i> Settings<'a, 't, 'i> {
localized_attributes_rules: Setting::NotSet,
prefix_search: Setting::NotSet,
facet_search: Setting::NotSet,
chat: Setting::NotSet,
indexer_config,
}
}
@@ -354,6 +359,14 @@ impl<'a, 't, 'i> Settings<'a, 't, 'i> {
self.min_word_len_one_typo = Setting::Reset;
}
pub fn set_disable_on_numbers(&mut self, disable_on_numbers: bool) {
self.disable_on_numbers = Setting::Set(disable_on_numbers);
}
pub fn reset_disable_on_numbers(&mut self) {
self.disable_on_numbers = Setting::Reset;
}
pub fn set_exact_words(&mut self, words: BTreeSet<String>) {
self.exact_words = Setting::Set(words);
}
@@ -442,6 +455,14 @@ impl<'a, 't, 'i> Settings<'a, 't, 'i> {
self.facet_search = Setting::Reset;
}
pub fn set_chat(&mut self, value: ChatSettings) {
self.chat = Setting::Set(value);
}
pub fn reset_chat(&mut self) {
self.chat = Setting::Reset;
}
#[tracing::instrument(
level = "trace"
skip(self, progress_callback, should_abort, settings_diff),
@@ -866,6 +887,24 @@ impl<'a, 't, 'i> Settings<'a, 't, 'i> {
Ok(())
}
fn update_disabled_typos_terms(&mut self) -> Result<()> {
let mut disabled_typos_terms = self.index.disabled_typos_terms(self.wtxn)?;
match self.disable_on_numbers {
Setting::Set(disable_on_numbers) => {
disabled_typos_terms.disable_on_numbers = disable_on_numbers;
}
Setting::Reset => {
self.index.delete_disabled_typos_terms(self.wtxn)?;
disabled_typos_terms.disable_on_numbers =
DisabledTyposTerms::default().disable_on_numbers;
}
Setting::NotSet => (),
}
self.index.put_disabled_typos_terms(self.wtxn, &disabled_typos_terms)?;
Ok(())
}
fn update_exact_words(&mut self) -> Result<()> {
match self.exact_words {
Setting::Set(ref mut words) => {
@@ -1210,6 +1249,45 @@ impl<'a, 't, 'i> Settings<'a, 't, 'i> {
Ok(())
}
fn update_chat_config(&mut self) -> heed::Result<bool> {
match &mut self.chat {
Setting::Set(ChatSettings {
description: new_description,
document_template: new_document_template,
document_template_max_bytes: new_document_template_max_bytes,
}) => {
let mut old = self.index.chat_config(self.wtxn)?;
let ChatConfig {
ref mut description,
prompt: PromptData { ref mut template, ref mut max_bytes },
} = old;
match new_description {
Setting::Set(d) => *description = d.clone(),
Setting::Reset => *description = Default::default(),
Setting::NotSet => (),
}
match new_document_template {
Setting::Set(dt) => *template = dt.clone(),
Setting::Reset => *template = Default::default(),
Setting::NotSet => (),
}
match new_document_template_max_bytes {
Setting::Set(m) => *max_bytes = NonZeroUsize::new(*m),
Setting::Reset => *max_bytes = Some(default_max_bytes()),
Setting::NotSet => (),
}
self.index.put_chat_config(self.wtxn, &old)?;
Ok(true)
}
Setting::Reset => self.index.delete_chat_config(self.wtxn),
Setting::NotSet => Ok(false),
}
}
pub fn execute<FP, FA>(mut self, progress_callback: FP, should_abort: FA) -> Result<()>
where
FP: Fn(UpdateIndexingStep) + Sync,
@@ -1246,6 +1324,8 @@ impl<'a, 't, 'i> Settings<'a, 't, 'i> {
self.update_prefix_search()?;
self.update_facet_search()?;
self.update_localized_attributes_rules()?;
self.update_disabled_typos_terms()?;
self.update_chat_config()?;
let embedding_config_updates = self.update_embedding_configs()?;
@@ -1327,6 +1407,7 @@ impl InnerIndexSettingsDiff {
|| old_settings.prefix_search != new_settings.prefix_search
|| old_settings.localized_attributes_rules
!= new_settings.localized_attributes_rules
|| old_settings.disabled_typos_terms != new_settings.disabled_typos_terms
};
let cache_exact_attributes = old_settings.exact_attributes != new_settings.exact_attributes;
@@ -1526,6 +1607,7 @@ pub(crate) struct InnerIndexSettings {
pub user_defined_searchable_attributes: Option<Vec<String>>,
pub sortable_fields: HashSet<String>,
pub exact_attributes: HashSet<FieldId>,
pub disabled_typos_terms: DisabledTyposTerms,
pub proximity_precision: ProximityPrecision,
pub embedding_configs: EmbeddingConfigs,
pub geo_fields_ids: Option<(FieldId, FieldId)>,
@@ -1574,7 +1656,7 @@ impl InnerIndexSettings {
.map(|fields| fields.into_iter().map(|f| f.to_string()).collect());
let builder = MetadataBuilder::from_index(index, rtxn)?;
let fields_ids_map = FieldIdMapWithMetadata::new(fields_ids_map, builder);
let disabled_typos_terms = index.disabled_typos_terms(rtxn)?;
Ok(Self {
stop_words,
allowed_separators,
@@ -1592,6 +1674,7 @@ impl InnerIndexSettings {
geo_fields_ids,
prefix_search,
facet_search,
disabled_typos_terms,
})
}

View File

@@ -896,6 +896,7 @@ fn test_correct_settings_init() {
localized_attributes_rules,
prefix_search,
facet_search,
disable_on_numbers,
} = settings;
assert!(matches!(searchable_fields, Setting::NotSet));
assert!(matches!(displayed_fields, Setting::NotSet));
@@ -923,6 +924,7 @@ fn test_correct_settings_init() {
assert!(matches!(localized_attributes_rules, Setting::NotSet));
assert!(matches!(prefix_search, Setting::NotSet));
assert!(matches!(facet_search, Setting::NotSet));
assert!(matches!(disable_on_numbers, Setting::NotSet));
})
.unwrap();
}

View File

@@ -1,12 +1,14 @@
mod v1_12;
mod v1_13;
mod v1_14;
mod v1_15;
use heed::RwTxn;
use v1_12::{V1_12_3_To_V1_13_0, V1_12_To_V1_12_3};
use v1_13::{V1_13_0_To_V1_13_1, V1_13_1_To_Latest_V1_13};
use v1_14::Latest_V1_13_To_Latest_V1_14;
use v1_15::Latest_V1_14_To_Latest_V1_15;
use crate::constants::{VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH};
use crate::progress::{Progress, VariableNameStep};
use crate::{Index, InternalError, Result};
@@ -23,12 +25,16 @@ trait UpgradeIndex {
}
/// Return true if the cached stats of the index must be regenerated
pub fn upgrade(
pub fn upgrade<MSP>(
wtxn: &mut RwTxn,
index: &Index,
db_version: (u32, u32, u32),
must_stop_processing: MSP,
progress: Progress,
) -> Result<bool> {
) -> Result<bool>
where
MSP: Fn() -> bool + Sync,
{
let from = index.get_version(wtxn)?.unwrap_or(db_version);
let upgrade_functions: &[&dyn UpgradeIndex] = &[
&V1_12_To_V1_12_3 {},
@@ -36,6 +42,10 @@ pub fn upgrade(
&V1_13_0_To_V1_13_1 {},
&V1_13_1_To_Latest_V1_13 {},
&Latest_V1_13_To_Latest_V1_14 {},
&Latest_V1_14_To_Latest_V1_15 {},
// This is the last upgrade function, it will be called when the index is up to date.
// any other upgrade function should be added before this one.
&ToCurrentNoOp {},
];
let start = match from {
@@ -43,8 +53,9 @@ pub fn upgrade(
(1, 12, 3..) => 1,
(1, 13, 0) => 2,
(1, 13, _) => 4,
(1, 14, _) => 5,
// We must handle the current version in the match because in case of a failure some index may have been upgraded but not other.
(1, 14, _) => 4,
(1, 15, _) => 6,
(major, minor, patch) => {
return Err(InternalError::CannotUpgradeToVersion(major, minor, patch).into())
}
@@ -56,6 +67,9 @@ pub fn upgrade(
let mut current_version = from;
let mut regenerate_stats = false;
for (i, upgrade) in upgrade_path.iter().enumerate() {
if (must_stop_processing)() {
return Err(crate::Error::InternalError(InternalError::AbortedIndexation));
}
let target = upgrade.target_version();
progress.update_progress(VariableNameStep::<UpgradeVersion>::new(
format!(
@@ -77,3 +91,22 @@ pub fn upgrade(
Ok(regenerate_stats)
}
#[allow(non_camel_case_types)]
struct ToCurrentNoOp {}
impl UpgradeIndex for ToCurrentNoOp {
fn upgrade(
&self,
_wtxn: &mut RwTxn,
_index: &Index,
_original: (u32, u32, u32),
_progress: Progress,
) -> Result<bool> {
Ok(false)
}
fn target_version(&self) -> (u32, u32, u32) {
(VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH)
}
}

View File

@@ -1,7 +1,6 @@
use heed::RwTxn;
use super::UpgradeIndex;
use crate::constants::{VERSION_MAJOR, VERSION_MINOR, VERSION_PATCH};
use crate::database_stats::DatabaseStats;
use crate::progress::Progress;
use crate::{make_enum_progress, Index, Result};
@@ -51,10 +50,6 @@ impl UpgradeIndex for V1_13_1_To_Latest_V1_13 {
}
fn target_version(&self) -> (u32, u32, u32) {
(
VERSION_MAJOR.parse().unwrap(),
VERSION_MINOR.parse().unwrap(),
VERSION_PATCH.parse().unwrap(),
)
(1, 13, 3)
}
}

View File

@@ -0,0 +1,35 @@
use heed::RwTxn;
use super::UpgradeIndex;
use crate::progress::Progress;
use crate::update::new::indexer::recompute_word_fst_from_word_docids_database;
use crate::{make_enum_progress, Index, Result};
#[allow(non_camel_case_types)]
pub(super) struct Latest_V1_14_To_Latest_V1_15();
impl UpgradeIndex for Latest_V1_14_To_Latest_V1_15 {
fn upgrade(
&self,
wtxn: &mut RwTxn,
index: &Index,
_original: (u32, u32, u32),
progress: Progress,
) -> Result<bool> {
// Recompute the word FST from the word docids database.
make_enum_progress! {
enum TypoTolerance {
RecomputeWordFst,
}
};
progress.update_progress(TypoTolerance::RecomputeWordFst);
recompute_word_fst_from_word_docids_database(index, wtxn)?;
Ok(false)
}
fn target_version(&self) -> (u32, u32, u32) {
(1, 15, 0)
}
}

View File

@@ -33,6 +33,7 @@ pub struct EmbeddingSettings {
///
/// - Defaults to `openAi`
pub source: Setting<EmbedderSource>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<String>)]
@@ -55,6 +56,7 @@ pub struct EmbeddingSettings {
/// - For source `openAi`, defaults to `text-embedding-3-small`
/// - For source `huggingFace`, defaults to `BAAI/bge-base-en-v1.5`
pub model: Setting<String>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<String>)]
@@ -75,6 +77,7 @@ pub struct EmbeddingSettings {
/// - When `model` is set to default, defaults to `617ca489d9e86b49b8167676d8220688b99db36e`
/// - Otherwise, defaults to `null`
pub revision: Setting<String>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<OverridePooling>)]
@@ -96,6 +99,7 @@ pub struct EmbeddingSettings {
///
/// - Embedders created before this parameter was available default to `forceMean` to preserve the existing behavior.
pub pooling: Setting<OverridePooling>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<String>)]
@@ -118,6 +122,7 @@ pub struct EmbeddingSettings {
///
/// - This setting is partially hidden when returned by the settings
pub api_key: Setting<String>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<String>)]
@@ -141,6 +146,7 @@ pub struct EmbeddingSettings {
/// - For source `openAi`, the dimensions is the maximum allowed by the model.
/// - For sources `ollama` and `rest`, the dimensions are inferred by embedding a sample text.
pub dimensions: Setting<usize>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<bool>)]
@@ -167,6 +173,7 @@ pub struct EmbeddingSettings {
/// first enabling it. If you are unsure of whether the performance-relevancy tradeoff is right for you,
/// we recommend to use this parameter on a test index first.
pub binary_quantized: Setting<bool>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<bool>)]
@@ -183,6 +190,7 @@ pub struct EmbeddingSettings {
///
/// - 🏗️ When modified, embeddings are regenerated for documents whose rendering through the template produces a different text.
pub document_template: Setting<String>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<usize>)]
@@ -201,6 +209,7 @@ pub struct EmbeddingSettings {
///
/// - Defaults to 400
pub document_template_max_bytes: Setting<usize>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<String>)]
@@ -219,6 +228,7 @@ pub struct EmbeddingSettings {
/// - 🌱 When modified for source `openAi`, embeddings are never regenerated
/// - 🏗️ When modified for sources `ollama` and `rest`, embeddings are always regenerated
pub url: Setting<String>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<serde_json::Value>)]
@@ -236,6 +246,7 @@ pub struct EmbeddingSettings {
///
/// - 🏗️ Changing the value of this parameter always regenerates embeddings
pub request: Setting<serde_json::Value>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<serde_json::Value>)]
@@ -253,6 +264,7 @@ pub struct EmbeddingSettings {
///
/// - 🏗️ Changing the value of this parameter always regenerates embeddings
pub response: Setting<serde_json::Value>,
#[serde(default, skip_serializing_if = "Setting::is_not_set")]
#[deserr(default)]
#[schema(value_type = Option<BTreeMap<String, String>>)]