Nov 28, 2025 · Nov 28, 2025 · Nov 28, 2025 · Nov 29, 2025
diff --git a/src/parser/transform/helpers/transform_sample.cpp b/src/parser/transform/helpers/transform_sample.cpp
 } else {
 // sample size is given in rows: use reservoir sampling
 auto rows = sample_value.GetValue<int64_t>();
 if (rows < 0) {
 throw ParserException("Sample rows %lld out of range, must be bigger than or equal to 0", rows);
 }
 if (rows >= Allocator::MAXIMUM_ALLOC_SIZE) {
 throw ParserException("Cannot sample over %d rows. Please use percentage instead",
 if (rows < 0 || sample_value.GetValue<uint64_t>() >= Allocator::MAXIMUM_ALLOC_SIZE) {
 throw ParserException("Sample rows %lld out of range, must be between 0 and %lld", rows,
                      Allocator::MAXIMUM_ALLOC_SIZE);
 }
 result->sample_size = Value::BIGINT(rows);
diff --git a/test/sql/sample/test_sample_too_big.test b/test/sql/sample/test_sample_too_big.test
 statement error
 SELECT * FROM t1 TABLESAMPLE RESERVOIR(1222222220022220);
 ----
 <REGEX>:.*Cannot sample over.*
 <REGEX>:.*Sample rows.*out of range.*

 statement error
 SELECT * FROM t1 WHERE a IN (SELECT * FROM t1 TABLESAMPLE RESERVOIR(1222222220022220));
 ----
 <REGEX>:.*Cannot sample over.*
 <REGEX>:.*Sample rows.*out of range.*
Original file line number	Diff line number	Diff line change
Expand Up		@@ -44,11 +44,8 @@ unique_ptr<SampleOptions> Transformer::TransformSampleOptions(optional_ptr<duckd
		} else {
		// sample size is given in rows: use reservoir sampling
		auto rows = sample_value.GetValue<int64_t>();
		if (rows < 0) {
		throw ParserException("Sample rows %lld out of range, must be bigger than or equal to 0", rows);
		}
		if (rows >= Allocator::MAXIMUM_ALLOC_SIZE) {
		throw ParserException("Cannot sample over %d rows. Please use percentage instead",
		if (rows < 0 \|\| sample_value.GetValue<uint64_t>() >= Allocator::MAXIMUM_ALLOC_SIZE) {
Copy link Collaborator MytherinNov 29, 2025• edited Loading Choose a reason for hiding this comment The reason will be displayed to describe this comment to others.Learn more. The sample size is in rows but the allocator max size is in bytes - I suspect that sampling for`281474976710656-1` rows will still throw the allocation error. Should we reduce the limit further / add a test for this? Alternatively, we might want to not allocate all the space for the sample up-front if the sample size is very large - but that's obviously a larger change.
		throw ParserException("Sample rows %lld out of range, must be between 0 and %lld", rows,
		Allocator::MAXIMUM_ALLOC_SIZE);
		}
		result->sample_size = Value::BIGINT(rows);
Expand Down
Original file line number	Diff line number	Diff line change
Expand Up		@@ -11,9 +11,9 @@ INSERT INTO t1 VALUES(1), (2), (3), (3), (5);
		statement error
		SELECT * FROM t1 TABLESAMPLE RESERVOIR(1222222220022220);
		----
		<REGEX>:.Cannot sample over.
		<REGEX>:.Sample rows.out of range.*

		statement error
		SELECT * FROM t1 WHERE a IN (SELECT * FROM t1 TABLESAMPLE RESERVOIR(1222222220022220));
		----
		<REGEX>:.Cannot sample over.
		<REGEX>:.Sample rows.out of range.*