Apache DataFusion SQL Query Engine
Adding Use of arrow's has_true() / has_false() (#21806)
Raushan Prabhakar9bbc28b8111e
diff --git a/datafusion/datasource-parquet/src/metadata.rs b/datafusion/datasource-parquet/src/metadata.rsindex 3d4d051abb..70ac3fe498 100644--- a/datafusion/datasource-parquet/src/metadata.rs+++ b/datafusion/datasource-parquet/src/metadata.rs@@ -520,12 +520,10 @@ fn summarize_column_statistics(// handle the common special case when all row groups have exact statisticslet exactness = &is_max_value_exact_stat;- if !exactness.is_empty()- && exactness.null_count() == 0- && exactness.true_count() == exactness.len()+ if !exactness.is_empty() && exactness.null_count() == 0 && !exactness.has_false(){accumulators.is_max_value_exact[logical_schema_index] = Some(true);- } else if exactness.true_count() == 0 {+ } else if !exactness.has_true() {accumulators.is_max_value_exact[logical_schema_index] = Some(false);} else {let val = max_acc.evaluate()?;@@ -539,12 +537,10 @@ fn summarize_column_statistics(// handle the common special case when all row groups have exact statisticslet exactness = &is_min_value_exact_stat;- if !exactness.is_empty()- && exactness.null_count() == 0- && exactness.true_count() == exactness.len()+ if !exactness.is_empty() && exactness.null_count() == 0 && !exactness.has_false(){accumulators.is_min_value_exact[logical_schema_index] = Some(true);- } else if exactness.true_count() == 0 {+ } else if !exactness.has_true() {accumulators.is_min_value_exact[logical_schema_index] = Some(false);} else {let val = min_acc.evaluate()?;@@ -675,7 +671,7 @@ fn has_any_exact_match(let scalar_array = value.to_scalar().ok()?;let eq_mask = eq(&scalar_array, &array).ok()?;let combined_mask = and(&eq_mask, exactness).ok()?;- Some(combined_mask.true_count() > 0)+ Some(combined_mask.has_true())}/// Wrapper to implement [`FileMetadata`] for [`ParquetMetaData`].diff --git a/datafusion/functions-nested/src/array_has.rs b/datafusion/functions-nested/src/array_has.rsindex d5b2e3e6cf..04818258f0 100644--- a/datafusion/functions-nested/src/array_has.rs+++ b/datafusion/functions-nested/src/array_has.rs@@ -338,7 +338,7 @@ fn array_has_dispatch_for_array<'a>(let is_nested = arr.data_type().is_nested();let needle_row = Scalar::new(needle.slice(i, 1));let eq_array = compare_with_eq(&arr, &needle_row, is_nested)?;- result.append(eq_array.true_count() > 0);+ result.append(eq_array.has_true());}Ok(Arc::new(BooleanArray::new(result.finish(), combined_nulls)))diff --git a/datafusion/functions-nested/src/replace.rs b/datafusion/functions-nested/src/replace.rsindex 3de66810fc..7effd13b69 100644--- a/datafusion/functions-nested/src/replace.rs+++ b/datafusion/functions-nested/src/replace.rs@@ -347,7 +347,7 @@ fn general_replace<O: OffsetSizeTrait>(let mut counter = 0;// All elements are false, no need to replace, just copy original data- if eq_array.false_count() == eq_array.len() {+ if !eq_array.has_true() {mutable.extend(original_idx.to_usize().unwrap(),start.to_usize().unwrap(),diff --git a/datafusion/physical-expr-common/src/physical_expr.rs b/datafusion/physical-expr-common/src/physical_expr.rsindex bdaddc324e..2fc3bf2afa 100644--- a/datafusion/physical-expr-common/src/physical_expr.rs+++ b/datafusion/physical-expr-common/src/physical_expr.rs@@ -23,7 +23,7 @@ use std::sync::Arc;use crate::utils::scatter;-use arrow::array::{ArrayRef, BooleanArray, new_empty_array};+use arrow::array::{Array, ArrayRef, BooleanArray, new_empty_array};use arrow::compute::filter_record_batch;use arrow::datatypes::{DataType, Field, FieldRef, Schema};use arrow::record_batch::RecordBatch;@@ -109,17 +109,15 @@ pub trait PhysicalExpr: Any + Send + Sync + Display + Debug + DynEq + DynHash {);}- let selection_count = selection.true_count();-// First, check if we can avoid filtering altogether.- if selection_count == row_count {+ if selection.null_count() == 0 && !selection.has_false() {// All values from the `selection` filter are true and match the input batch.// No need to perform any filtering.return self.evaluate(batch);}// Next, prepare the result array for each 'true' row in the selection vector.- let filtered_result = if selection_count == 0 {+ let filtered_result = if !selection.has_true() {// Do not call `evaluate` when the selection is empty.// `evaluate_selection` is used to conditionally evaluate expressions.// When the expression in question is fallible, evaluating it with an empty@@ -132,7 +130,7 @@ pub trait PhysicalExpr: Any + Send + Sync + Display + Debug + DynEq + DynHash {// If we reach this point, there's no other option than to filter the batch.// This is a fairly costly operation since it requires creating partial copies// (worst case of length `row_count - 1`) of all the arrays in the record batch.- // The resulting `filtered_batch` will contain `selection_count` rows.+ // The resulting `filtered_batch` will contain one row per true in `selection`.let filtered_batch = filter_record_batch(batch, selection)?;self.evaluate(&filtered_batch)?};diff --git a/datafusion/physical-expr-common/src/utils.rs b/datafusion/physical-expr-common/src/utils.rsindex 6ca00a4460..e469885f83 100644--- a/datafusion/physical-expr-common/src/utils.rs+++ b/datafusion/physical-expr-common/src/utils.rs@@ -83,18 +83,18 @@ pub fn scatter(mask: &BooleanArray, truthy: &dyn Array) -> Result<ArrayRef> {};let output_len = mask.len();- let count = mask.true_count();// Fast path: no true values mean all-null object- if count == 0 {+ if !mask.has_true() {return Ok(new_null_array(truthy.data_type(), output_len));}// Fast path: all true means output = truthy- if count == output_len {+ if mask.null_count() == 0 && !mask.has_false() {return Ok(truthy.slice(0, truthy.len()));}+ let count = mask.true_count();let selectivity = count as f64 / output_len as f64;let mask_buffer = mask.values();diff --git a/datafusion/physical-expr/src/expressions/binary.rs b/datafusion/physical-expr/src/expressions/binary.rsindex 268a350a4d..ae93b1d03a 100644--- a/datafusion/physical-expr/src/expressions/binary.rs+++ b/datafusion/physical-expr/src/expressions/binary.rs@@ -289,11 +289,11 @@ impl PhysicalExpr for BinaryExpr {ColumnarValue::Array(array) => {// When the array on the right is all true or all false, skip the scatter processlet boolean_array = array.as_boolean();- let true_count = boolean_array.true_count();- let length = boolean_array.len();- if true_count == length {+ if boolean_array.null_count() == 0 && !boolean_array.has_false() {return Ok(lhs);- } else if true_count == 0 && boolean_array.null_count() == 0 {+ } else if boolean_array.null_count() == 0+ && !boolean_array.has_true()+ {// If the right-hand array is returned at this point,the lengths will be inconsistent;// returning a scalar can avoid this issuereturn Ok(ColumnarValue::Scalar(ScalarValue::Boolean(diff --git a/datafusion/physical-expr/src/expressions/case.rs b/datafusion/physical-expr/src/expressions/case.rsindex bf95f8e6ac..919b6684e0 100644--- a/datafusion/physical-expr/src/expressions/case.rs+++ b/datafusion/physical-expr/src/expressions/case.rs@@ -814,17 +814,15 @@ impl CaseBody {}}?;- // `true_count` ignores `true` values where the validity bit is not set, so there's- // no need to call `prep_null_mask_filter`.- let when_true_count = when_value.true_count();-- // If the 'when' predicate did not match any rows, continue to the next branch immediately- if when_true_count == 0 {+ // If the 'when' predicate did not match any rows, continue to the next branch immediately.+ // Only counts valid slots that are true (masked-null predicate slots are ignored),+ // so no `prep_null_mask_filter` needed here.+ if !when_value.has_true() {continue;}// If the 'when' predicate matched all remaining rows, there is no need to filter- if when_true_count == remainder_batch.num_rows() {+ if when_value.null_count() == 0 && !when_value.has_false() {let then_expression = &self.when_then_expr[i].1;let then_value = then_expression.evaluate(&remainder_batch)?;result_builder.add_branch_result(&remainder_rows, then_value)?;@@ -903,17 +901,15 @@ impl CaseBody {internal_datafusion_err!("WHEN expression did not return a BooleanArray")})?;- // `true_count` ignores `true` values where the validity bit is not set, so there's- // no need to call `prep_null_mask_filter`.- let when_true_count = when_value.true_count();-- // If the 'when' predicate did not match any rows, continue to the next branch immediately- if when_true_count == 0 {+ // If the 'when' predicate did not match any rows, continue to the next branch immediately.+ // Only counts valid slots that are true (masked-null predicate slots are ignored)+ // so no `prep_null_mask_filter` needed here.+ if !when_value.has_true() {continue;}// If the 'when' predicate matched all remaining rows, there is no need to filter- if when_true_count == remainder_batch.num_rows() {+ if when_value.null_count() == 0 && !when_value.has_false() {let then_expression = &self.when_then_expr[i].1;let then_value = then_expression.evaluate(&remainder_batch)?;result_builder.add_branch_result(&remainder_rows, then_value)?;@@ -1165,11 +1161,10 @@ impl CaseExpr {)})?;- let true_count = when_value.true_count();- if true_count == when_value.len() {+ if when_value.null_count() == 0 && !when_value.has_false() {// All input rows are true, just call the 'then' expressionself.body.when_then_expr[0].1.evaluate(batch)- } else if true_count == 0 {+ } else if !when_value.has_true() {// All input rows are false/null, just call the 'else' expressionmatch &self.body.else_expr {Some(else_expr) => else_expr.evaluate(batch),diff --git a/datafusion/physical-expr/src/expressions/in_list.rs b/datafusion/physical-expr/src/expressions/in_list.rsindex ede493dcf1..44e6a99c8a 100644--- a/datafusion/physical-expr/src/expressions/in_list.rs+++ b/datafusion/physical-expr/src/expressions/in_list.rs@@ -377,7 +377,7 @@ impl PhysicalExpr for InListExpr {for expr in self.list.iter().skip(1) {// Short-circuit: if every non-null row is already true,// no further list items can change the result.- if found.null_count() == 0 && found.true_count() == num_rows {+ if found.null_count() == 0 && !found.has_false() {break;}found = or_kleene(&found, &compare_one(expr)?)?;diff --git a/datafusion/physical-plan/src/joins/nested_loop_join.rs b/datafusion/physical-plan/src/joins/nested_loop_join.rsindex c686d9d5a0..add4508c87 100644--- a/datafusion/physical-plan/src/joins/nested_loop_join.rs+++ b/datafusion/physical-plan/src/joins/nested_loop_join.rs@@ -2105,7 +2105,7 @@ impl NestedLoopJoinStream {return Ok(None);}- if cur_right_bitmap.true_count() == 0 {+ if !cur_right_bitmap.has_true() {// If none of the pairs has passed the join predicate/filterOk(None)} else {@@ -2300,11 +2300,8 @@ impl NestedLoopJoinStream {) -> Result<()> {let left_data = self.get_left_data()?;- // number of successfully joined pairs from (l_index x cur_right_batch)- let joined_len = r_matched_bitmap.true_count();-// 1. Maybe update the left bitmap- if need_produce_result_in_final(self.join_type) && (joined_len > 0) {+ if need_produce_result_in_final(self.join_type) && r_matched_bitmap.has_true() {let mut bitmap = left_data.bitmap().lock();bitmap.set_bit(l_index, true);}@@ -2696,7 +2693,7 @@ fn build_unmatched_batch(not(&batch_bitmap)?};- if bitmap.true_count() == 0 {+ if !bitmap.has_true() {return Ok(None);}diff --git a/datafusion/physical-plan/src/joins/sort_merge_join/filter.rs b/datafusion/physical-plan/src/joins/sort_merge_join/filter.rsindex 3a5f0d0f05..4fc6cccaa8 100644--- a/datafusion/physical-plan/src/joins/sort_merge_join/filter.rs+++ b/datafusion/physical-plan/src/joins/sort_merge_join/filter.rs@@ -331,7 +331,7 @@ pub fn filter_record_batch_by_join_type(.unwrap();// All rows passed the filter — no null-joining needed- if kept_corrected.true_count() == kept_corrected.len() {+ if !kept_corrected.has_false() {return Ok(kept_batch);}diff --git a/datafusion/physical-plan/src/topk/mod.rs b/datafusion/physical-plan/src/topk/mod.rsindex 2c6d3bcad1..15c9f0a060 100644--- a/datafusion/physical-plan/src/topk/mod.rs+++ b/datafusion/physical-plan/src/topk/mod.rs@@ -254,13 +254,12 @@ impl TopK {let num_rows = batch.num_rows();let array = filtered.into_array(num_rows)?;let mut filter = array.as_boolean().clone();- let true_count = filter.true_count();- if true_count == 0 {+ if !filter.has_true() {// nothing to filter, so no need to updatereturn Ok(());}// only update the keys / rows if the filter does not match all rows- if true_count < num_rows {+ if filter.null_count() > 0 || filter.has_false() {// Indices in `set_indices` should be correct if filter contains nulls// So we prepare the filter here. Note this is also done in the `FilterBuilder`// so there is no overhead to do this here.diff --git a/datafusion/spark/src/function/array/array_contains.rs b/datafusion/spark/src/function/array/array_contains.rsindex e0f99f7dd3..5c7cb4be6f 100644--- a/datafusion/spark/src/function/array/array_contains.rs+++ b/datafusion/spark/src/function/array/array_contains.rs@@ -84,7 +84,7 @@ fn apply_spark_null_semantics(haystack_arg: &ColumnarValue,) -> Result<BooleanArray> {// happy path- if result.false_count() == 0 || haystack_arg.data_type() == DataType::Null {+ if haystack_arg.data_type() == DataType::Null || !result.has_false() {return Ok(result.clone());}