refactor code

yanghua · yanghua · commit 8d896287fddc · 2025-09-30T16:31:53.000+08:00
diff --git a/rust/lance/src/dataset/conflict_detection/conflict_detector.rs b/rust/lance/src/dataset/conflict_detection/conflict_detector.rs
@@ -256,8 +256,8 @@ mod tests {
         let mut filter2 = PrimaryKeyBloomFilter::new(vec!["id".to_string()]);
 
         let shared_key = PrimaryKeyValue::String("shared_key".to_string());
-        filter1.insert(shared_key.clone(), 1).unwrap();
-        filter2.insert(shared_key, 2).unwrap();
+        filter1.insert(shared_key.clone()).unwrap();
+        filter2.insert(shared_key).unwrap();
 
         let result = detector
             .check_filter_conflict(&filter1, &filter2, "uuid2", 2)
@@ -272,15 +272,15 @@ mod tests {
         let detector = DefaultConflictDetector::new();
         let mut current_filter = PrimaryKeyBloomFilter::new(vec!["id".to_string()]);
         let shared_key = PrimaryKeyValue::String("shared_key".to_string());
-        current_filter.insert(shared_key.clone(), 1).unwrap();
+        current_filter.insert(shared_key.clone()).unwrap();
 
         // Create concurrent transactions
         let mut filter1 = PrimaryKeyBloomFilter::new(vec!["id".to_string()]);
-        filter1.insert(shared_key.clone(), 2).unwrap();
+        filter1.insert(shared_key.clone()).unwrap();
 
         let mut filter2 = PrimaryKeyBloomFilter::new(vec!["id".to_string()]);
         filter2
-            .insert(PrimaryKeyValue::String("other_key".to_string()), 3)
+            .insert(PrimaryKeyValue::String("other_key".to_string()))
             .unwrap();
 
         let transactions = vec![
diff --git a/rust/lance/src/dataset/conflict_detection/primary_key_filter.rs b/rust/lance/src/dataset/conflict_detection/primary_key_filter.rs
@@ -140,12 +140,6 @@ impl PrimaryKeyBloomFilter {
 
 #[cfg(test)]
 mod tests {
-    use super::*;
-
-    use arrow_array::{Int64Array, RecordBatch, StringArray};
-    use arrow_schema::{DataType, Field, Schema};
-    use std::sync::Arc;
-
     use crate::dataset::conflict_detection::{
         conflict_detector::{ConflictDetector, DefaultConflictDetector},
         primary_key_filter::{PrimaryKeyBloomFilter, PrimaryKeyValue},
@@ -193,7 +187,7 @@ mod tests {
         filter2.insert(key3).unwrap();
 
         // Should detect intersection
-        assert!(filter1.has_intersection(&filter2));
+        assert!(filter1.has_intersection(filter2));
     }
 
     #[test]
@@ -213,18 +207,6 @@ mod tests {
         );
     }
 
-    fn create_test_record_batch(user_ids: Vec<&str>, scores: Vec<i64>) -> RecordBatch {
-        let schema = Arc::new(Schema::new(vec![
-            Field::new("user_id", DataType::Utf8, false),
-            Field::new("score", DataType::Int64, false),
-        ]));
-
-        let user_id_array = StringArray::from(user_ids);
-        let score_array = Int64Array::from(scores);
-
-        RecordBatch::try_new(schema, vec![Arc::new(user_id_array), Arc::new(score_array)]).unwrap()
-    }
-
     #[test]
     fn test_bloom_filter_creation_and_basic_operations() {
         let mut bloom_filter = PrimaryKeyBloomFilter::new(vec!["user_id".to_string()]);
@@ -283,7 +265,7 @@ mod tests {
             PrimaryKeyValue::Int64(67890),
         ];
 
-        for (idx, key) in test_keys.iter().enumerate() {
+        for (_idx, key) in test_keys.iter().enumerate() {
             original_filter.insert(key.clone()).unwrap();
         }
 
@@ -317,7 +299,7 @@ mod tests {
             PrimaryKeyValue::String("charlie".to_string()),
         ];
 
-        for (idx, key) in keys1.iter().enumerate() {
+        for (_idx, key) in keys1.iter().enumerate() {
             filter1.insert(key.clone()).unwrap();
         }
 
@@ -327,7 +309,7 @@ mod tests {
             PrimaryKeyValue::String("eve".to_string()),
         ];
 
-        for (idx, key) in keys2.iter().enumerate() {
+        for (_idx, key) in keys2.iter().enumerate() {
             filter2.insert(key.clone()).unwrap();
         }
 
@@ -338,7 +320,7 @@ mod tests {
         assert!(conflict_result.has_conflict(), "should detect conflict");
         assert_eq!(
             conflict_result.conflicting_uuid(),
-            Some(&"test_transaction_uuid".to_string())
+            Some(&"test_transaction_uuid".to_string()).map(|x| x.as_str())
         );
     }
 
@@ -355,7 +337,7 @@ mod tests {
             PrimaryKeyValue::String("charlie".to_string()),
         ];
 
-        for (idx, key) in keys1.iter().enumerate() {
+        for (_idx, key) in keys1.iter().enumerate() {
             filter1.insert(key.clone()).unwrap();
         }
 
@@ -365,7 +347,7 @@ mod tests {
             PrimaryKeyValue::String("frank".to_string()),
         ];
 
-        for (idx, key) in keys2.iter().enumerate() {
+        for (_idx, key) in keys2.iter().enumerate() {
             filter2.insert(key.clone()).unwrap();
         }
 
@@ -388,7 +370,7 @@ mod tests {
             PrimaryKeyValue::String("transaction_user_2".to_string()),
         ];
 
-        for (idx, key) in keys.iter().enumerate() {
+        for (_idx, key) in keys.iter().enumerate() {
             bloom_filter.insert(key.clone()).unwrap();
         }
 
@@ -410,7 +392,7 @@ mod tests {
 
     #[test]
     fn test_transaction_protobuf_serialization_with_bloom_filter() {
-        use crate::dataset::transaction::pb;
+        use lance_table::format::pb;
 
         let mut bloom_filter = PrimaryKeyBloomFilter::new(vec!["user_id".to_string()]);
         bloom_filter
@@ -431,8 +413,8 @@ mod tests {
         let pb_transaction: pb::Transaction = (&original_transaction).into();
         assert!(pb_transaction.primary_key_bloom_filter.is_some());
         assert_eq!(
-            pb_transaction.primary_key_bloom_filter.unwrap(),
-            filter_data
+            pb_transaction.primary_key_bloom_filter.as_ref().unwrap(),
+            &filter_data
         );
 
         let deserialized_transaction: Transaction = pb_transaction.try_into().unwrap();
@@ -557,7 +539,10 @@ mod tests {
                 conflict_result.has_conflict(),
                 "A conflict should be detected"
             );
-            assert_eq!(conflict_result.conflicting_uuid(), Some(&transaction2.uuid));
+            assert_eq!(
+                conflict_result.conflicting_uuid(),
+                Some(&transaction2.uuid).map(|x| x.as_str())
+            );
             // Since the current implementation uses an exact map, there are no false positives.
             assert!(!conflict_result.might_be_false_positive());
         } else {
diff --git a/rust/lance/src/dataset/write/commit.rs b/rust/lance/src/dataset/write/commit.rs
@@ -529,6 +529,7 @@ mod tests {
             blobs_op: None,
             tag: None,
             transaction_properties: None,
+            primary_key_bloom_filter: None,
         }
     }
 
@@ -804,6 +805,7 @@ mod tests {
             blobs_op: None,
             tag: None,
             transaction_properties: None,
+            primary_key_bloom_filter: None,
         };
         let res = CommitBuilder::new(dataset.clone())
             .execute_batch(vec![update_transaction])

Original file line number	Diff line number	Diff line change
`@@ -529,6 +529,7 @@ mod tests {`
`529`	`529`	`blobs_op: None,`
`530`	`530`	`tag: None,`
`531`	`531`	`transaction_properties: None,`
	`532`	`+ primary_key_bloom_filter: None,`
`532`	`533`	`}`
`533`	`534`	`}`
`534`	`535`
`@@ -804,6 +805,7 @@ mod tests {`
`804`	`805`	`blobs_op: None,`
`805`	`806`	`tag: None,`
`806`	`807`	`transaction_properties: None,`
	`808`	`+ primary_key_bloom_filter: None,`
`807`	`809`	`};`
`808`	`810`	`let res = CommitBuilder::new(dataset.clone())`
`809`	`811`	`.execute_batch(vec![update_transaction])`