chore: codefmt

Signed-off-by: Kould <[email protected]>

Author: KKould

src/query/service/tests/it/storages/fuse/bloom_index_meta_size.rs

@@ -334,6 +334,7 @@ fn build_test_segment_info(
         bloom_filter_index_location: Some(location_gen.block_bloom_index_location(&block_uuid)),
         bloom_filter_index_size: 0,
         inverted_index_size: None,
+        ngram_filter_index_size: None,
         virtual_block_meta: None,
         compression: Compression::Lz4,
         create_on: Some(Utc::now()),

src/query/service/tests/it/storages/fuse/operations/mutation/recluster_mutator.rs

@@ -77,6 +77,7 @@ async fn test_recluster_mutator_block_select() -> Result<()> {
             0,
             None,
             None,
+            None,
             meta::Compression::Lz4Raw,
             Some(Utc::now()),
         ));

src/query/service/tests/it/storages/fuse/operations/mutation/segments_compact_mutator.rs

@@ -773,6 +773,7 @@ impl CompactSegmentTestFixture {
                         0,
                         None,
                         None,
+                        None,
                         Compression::Lz4Raw,
                         Some(Utc::now()),
                     );

src/query/service/tests/it/storages/fuse/operations/read_plan.rs

@@ -104,6 +104,7 @@ fn test_to_partitions() -> Result<()> {
         bloom_filter_size,
         None,
         None,
+        None,
         meta::Compression::Lz4Raw,
         Some(Utc::now()),
     ));

src/query/service/tests/it/storages/fuse/statistics.rs

@@ -628,6 +628,7 @@ fn test_reduce_block_meta() -> databend_common_exception::Result<()> {
             bloom_filter_index_size,
             None,
             None,
+            None,
             Compression::Lz4Raw,
             Some(Utc::now()),
         );

src/query/storages/common/index/src/bloom_index.rs

@@ -840,44 +840,42 @@ where T: EqVisitor
         let mut result = ControlFlow::Continue(None);
 
         if id.name() == "like" {
-            match args.as_slice() {
-                // patterns like `Column like <constant>`
-                [Expr::ColumnRef(ColumnRef {
-                    id,
-                    data_type: column_type,
-                    ..
-                }), Expr::Constant(Constant { scalar, .. })] => {
-                    if let Some(pattern) = scalar.as_string() {
-                        match generate_like_pattern(pattern.as_bytes(), 1) {
-                            LikePattern::StartOfPercent(v) | LikePattern::EndOfPercent(v) => {
-                                let string = String::from_utf8_lossy(v.as_ref()).to_string();
-
-                                result = self.0.enter_target(
-                                    *span,
-                                    id,
-                                    &Scalar::String(string),
-                                    column_type,
-                                    return_type,
-                                    true,
-                                )?;
-                            }
-                            LikePattern::SurroundByPercent(v) => {
-                                let string = String::from_utf8_lossy(v.needle()).to_string();
-
-                                result = self.0.enter_target(
-                                    *span,
-                                    id,
-                                    &Scalar::String(string),
-                                    column_type,
-                                    return_type,
-                                    true,
-                                )?;
-                            }
-                            _ => (),
+            // patterns like `Column like <constant>`
+            if let [Expr::ColumnRef(ColumnRef {
+                id,
+                data_type: column_type,
+                ..
+            }), Expr::Constant(Constant { scalar, .. })] = args.as_slice()
+            {
+                if let Some(pattern) = scalar.as_string() {
+                    match generate_like_pattern(pattern.as_bytes(), 1) {
+                        LikePattern::StartOfPercent(v) | LikePattern::EndOfPercent(v) => {
+                            let string = String::from_utf8_lossy(v.as_ref()).to_string();
+
+                            result = self.0.enter_target(
+                                *span,
+                                id,
+                                &Scalar::String(string),
+                                column_type,
+                                return_type,
+                                true,
+                            )?;
+                        }
+                        LikePattern::SurroundByPercent(v) => {
+                            let string = String::from_utf8_lossy(v.needle()).to_string();
+
+                            result = self.0.enter_target(
+                                *span,
+                                id,
+                                &Scalar::String(string),
+                                column_type,
+                                return_type,
+                                true,
+                            )?;
                         }
+                        _ => (),
                     }
                 }
-                _ => (),
             }
         } else {
             result = match args.as_slice() {

src/query/storages/common/index/src/filters/xor8/bloom_filter.rs

@@ -12,6 +12,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use std::cmp::max;
 use std::collections::HashSet;
 use std::hash::Hash;
 use std::hash::Hasher;
@@ -200,9 +201,9 @@ impl Filter for BloomFilter {
 }
 
 impl BloomFilter {
-    pub fn with_item_count(filter_size: usize, item_count: usize, seed: u64) -> Self {
+    pub fn with_item_count(filter_size: usize, mut item_count: usize, seed: u64) -> Self {
         assert!(filter_size > 0, "filter_size must be > 0");
-        assert!(item_count > 0, "item_count must be > 0");
+        item_count = max(item_count, 1);
 
         let ln2 = std::f64::consts::LN_2;
         let k = ((filter_size as f64 / item_count as f64) * ln2).ceil() as usize;
@@ -214,8 +215,7 @@ impl BloomFilter {
     pub fn with_params(size: usize, hashes: usize, seed: u64) -> Self {
         assert_ne!(size, 0);
         assert_ne!(hashes, 0);
-        let words =
-            (size + std::mem::size_of::<UnderType>() - 1) / std::mem::size_of::<UnderType>();
+        let words = size.div_ceil(std::mem::size_of::<UnderType>());
         Self {
             size,
             hashes,
@@ -227,8 +227,7 @@ impl BloomFilter {
 
     pub fn resize(&mut self, size: usize) {
         self.size = size;
-        self.words =
-            (size + std::mem::size_of::<UnderType>() - 1) / std::mem::size_of::<UnderType>();
+        self.words = size.div_ceil(std::mem::size_of::<UnderType>());
         self.filter.resize(self.words, 0);
     }
 

tests/sqllogictests/suites/mode/standalone/ee/explain_ngram_index.test

@@ -245,7 +245,7 @@ TableScan
 ├── table: default.test_ngram_index_db.t2
 ├── output columns: [id (#0), content (#1)]
 ├── read rows: 16
-├── read size: 1.20 KiB
+├── read size: 1.26 KiB
 ├── partitions total: 8
 ├── partitions scanned: 8
 ├── pruning stats: [segments: <range pruning: 1 to 1>, blocks: <range pruning: 8 to 8, bloom pruning: 8 to 8>]
@@ -259,7 +259,7 @@ TableScan
 ├── table: default.test_ngram_index_db.t2
 ├── output columns: [id (#0), content (#1)]
 ├── read rows: 16
-├── read size: 1.20 KiB
+├── read size: 1.26 KiB
 ├── partitions total: 8
 ├── partitions scanned: 8
 ├── pruning stats: [segments: <range pruning: 1 to 1>, blocks: <range pruning: 8 to 8, bloom pruning: 8 to 8>]