update cellulite to the latest version

crates/filter-parser/src/lib.rs

@@ -191,6 +191,7 @@ impl<'a> FilterCondition<'a> {
             FilterCondition::VectorExists { .. }
             | FilterCondition::GeoLowerThan { .. }
             | FilterCondition::GeoBoundingBox { .. }
+            | FilterCondition::GeoPolygon { .. }
             | FilterCondition::In { .. } => None,
         }
     }

crates/index-scheduler/src/error.rs

@@ -45,6 +45,7 @@ impl From<DateField> for Code {
     }
 }
 
+#[allow(clippy::large_enum_variant)]
 #[derive(Error, Debug)]
 pub enum Error {
     #[error("{1}")]

crates/meilisearch-types/src/lib.rs

@@ -1,3 +1,5 @@
+#![allow(clippy::result_large_err)]
+
 pub mod batch_view;
 pub mod batches;
 pub mod compression;

crates/meilitool/src/main.rs

@@ -1,3 +1,5 @@
+#![allow(clippy::result_large_err)]
+
 use std::fs::{read_dir, read_to_string, remove_file, File};
 use std::io::{BufWriter, Write as _};
 use std::path::PathBuf;

crates/milli/src/documents/primary_key.rs

@@ -48,6 +48,7 @@ pub enum PrimaryKey<'a> {
     Nested { name: &'a str },
 }
 
+#[allow(clippy::large_enum_variant)]
 pub enum DocumentIdExtractionError {
     InvalidDocumentId(UserError),
     MissingDocumentId,

crates/milli/src/error.rs

@@ -21,6 +21,7 @@ pub fn is_reserved_keyword(keyword: &str) -> bool {
         .contains(&keyword)
 }
 
+#[allow(clippy::large_enum_variant)]
 #[derive(Error, Debug)]
 pub enum Error {
     #[error("internal: {0}.")]
@@ -115,6 +116,7 @@ pub enum FieldIdMapMissingEntry {
     FieldName { field_name: String, process: &'static str },
 }
 
+#[allow(clippy::large_enum_variant)]
 #[derive(Error, Debug)]
 pub enum UserError {
     #[error(transparent)]

crates/milli/src/index.rs

@@ -244,7 +244,7 @@ impl Index {
         let embedder_category_id =
             env.create_database(&mut wtxn, Some(VECTOR_EMBEDDER_CATEGORY_ID))?;
         let vector_store = env.create_database(&mut wtxn, Some(VECTOR_STORE))?;
-        let cellulite = cellulite::Cellulite::create_from_env(&env, &mut wtxn)?;
+        let cellulite = cellulite::Cellulite::create_from_env(&env, &mut wtxn, "cellulite")?;
 
         let documents = env.create_database(&mut wtxn, Some(DOCUMENTS))?;
 
@@ -1972,13 +1972,17 @@ impl Index {
 
         // Cellulite
         const _CELLULITE_DB_CHECK: () = {
-            if Cellulite::nb_dbs() != 3 {
+            if Cellulite::nb_dbs() != 4 {
                 panic!("Cellulite database count has changed, please update the code accordingly.")
             }
         };
         sizes.insert("cellulite_item", self.cellulite.item_db_stats(rtxn).map(compute_size)?);
         sizes.insert("cellulite_cell", self.cellulite.cell_db_stats(rtxn).map(compute_size)?);
         sizes.insert("cellulite_update", self.cellulite.update_db_stats(rtxn).map(compute_size)?);
+        sizes.insert(
+            "cellulite_metadata",
+            self.cellulite.metadata_db_stats(rtxn).map(compute_size)?,
+        );
 
         Ok(sizes)
     }

crates/milli/src/progress.rs

@@ -278,30 +278,6 @@ impl<U: Send + Sync + 'static> Step for VariableNameStep<U> {
     }
 }
 
-// Integration with steppe
-
-impl steppe::Progress for Progress {
-    fn update(&self, sub_progress: impl steppe::Step) {
-        self.update_progress(Compat(sub_progress));
-    }
-}
-
-struct Compat<T: steppe::Step>(T);
-
-impl<T: steppe::Step> Step for Compat<T> {
-    fn name(&self) -> Cow<'static, str> {
-        self.0.name()
-    }
-
-    fn current(&self) -> u32 {
-        self.0.current().try_into().unwrap_or(u32::MAX)
-    }
-
-    fn total(&self) -> u32 {
-        self.0.total().try_into().unwrap_or(u32::MAX)
-    }
-}
-
 impl Step for arroy::MainStep {
     fn name(&self) -> Cow<'static, str> {
         match self {

crates/milli/src/search/facet/filter.rs

@@ -872,10 +872,10 @@ impl<'a> Filter<'a> {
 
                     let result = index
                         .cellulite
-                        .in_shape(rtxn, &polygon, &mut |_| ())
-                        .map_err(InternalError::CelluliteError)?;
+                        .in_shape(rtxn, &polygon)
+                        .map_err(InternalError::CelluliteError)?; // TODO: error code in invalid
 
-                    r2 = Some(RoaringBitmap::from_iter(result)); // TODO: Remove once we update roaring
+                    r2 = Some(RoaringBitmap::from_iter(result)); // TODO: Remove once we update roaring in meilisearch
                 }
 
                 match (r1, r2) {
@@ -924,10 +924,10 @@ impl<'a> Filter<'a> {
                 let polygon = geo_types::Polygon::new(geo_types::LineString(coords), Vec::new());
                 let result = index
                     .cellulite
-                    .in_shape(rtxn, &polygon, &mut |_| ())
-                    .map_err(InternalError::CelluliteError)?;
+                    .in_shape(rtxn, &polygon)
+                    .map_err(InternalError::CelluliteError)?; // TODO: update error code
 
-                let result = roaring::RoaringBitmap::from_iter(result); // TODO: Remove once we update roaring
+                let result = roaring::RoaringBitmap::from_iter(result); // TODO: Remove once we update roaring in meilisearch
 
                 Ok(result)
             }

crates/milli/src/update/index_documents/extract/mod.rs

@@ -230,6 +230,7 @@ pub fn request_threads() -> &'static ThreadPoolNoAbort {
 
 /// Extract chunked data and send it into lmdb_writer_sx sender:
 /// - documents
+#[allow(clippy::too_many_arguments)]
 fn send_original_documents_data(
     original_documents_chunk: Result<grenad::Reader<BufReader<File>>>,
     indexer: GrenadParameters,

crates/milli/src/update/index_documents/mod.rs

@@ -523,7 +523,7 @@ where
                 .is_some_and(|conf| conf.is_quantized);
             let is_quantizing = embedder_config.is_some_and(|action| action.is_being_quantized);
 
-            pool.install(|| {
+            pool.install(|| -> Result<_> {
                 let mut writer =
                     VectorStore::new(backend, vector_store, embedder_index, was_quantized);
                 writer.build_and_quantize(
@@ -541,7 +541,7 @@ where
             .map_err(InternalError::from)??;
         }
 
-        self.index.cellulite.build(self.wtxn, &Progress::default())?;
+        self.index.cellulite.build(self.wtxn, &self.should_abort, &Progress::default())?;
 
         self.execute_prefix_databases(
             word_docids.map(MergerBuilder::build),

crates/milli/src/update/new/indexer/mod.rs

@@ -189,7 +189,11 @@ where
 
         println!("Building geojson");
         indexing_context.progress.update_progress(IndexingStep::BuildingGeoJson);
-        index.cellulite.build(wtxn, indexing_context.progress)?;
+        index.cellulite.build(
+            wtxn,
+            &indexing_context.must_stop_processing,
+            indexing_context.progress,
+        )?;
 
         indexing_context.progress.update_progress(IndexingStep::Finalizing);
 
@@ -322,7 +326,11 @@ where
         .unwrap()?;
 
         indexing_context.progress.update_progress(IndexingStep::BuildingGeoJson);
-        index.cellulite.build(wtxn, indexing_context.progress)?;
+        index.cellulite.build(
+            wtxn,
+            &indexing_context.must_stop_processing,
+            indexing_context.progress,
+        )?;
 
         indexing_context.progress.update_progress(IndexingStep::Finalizing);
 

crates/milli/src/update/new/indexer/write.rs

@@ -17,7 +17,6 @@ use crate::vector::db::IndexEmbeddingConfig;
 use crate::vector::settings::EmbedderAction;
 use crate::vector::{Embedder, Embeddings, RuntimeEmbedders, VectorStore};
 use crate::{DocumentId, Error, Index, InternalError, Result, UserError};
-use crate::{Error, Index, InternalError, Result, UserError};
 
 pub fn write_to_db(
     mut writer_receiver: WriterBbqueueReceiver<'_>,