This is an automated email from the ASF dual-hosted git repository.
morningman pushed a commit to branch branch-1.2-lts
in repository https://gitbox.apache.org/repos/asf/doris.git
commit 683262c3c18b836bb21a78a0a7398e9ef0993d9c
Author: Ashin Gau <ashin...@users.noreply.github.com>
AuthorDate: Mon Dec 26 22:49:35 2022 +0800
[fix](multi-catalog) throw NPE when reading data after EOF (#15358)
1. Fix 1 bug:
Throw null pointer exception when reading data after the reader reaches the
end of file, so should return directly when `_do_lazy_read` read no data.
2. Optimize code:
Remove unused parameters.
3. Fix regression test
---
be/src/vec/exec/format/parquet/vparquet_column_reader.cpp | 1 -
be/src/vec/exec/format/parquet/vparquet_column_reader.h | 4 ++--
be/src/vec/exec/format/parquet/vparquet_group_reader.cpp | 3 ++-
3 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/be/src/vec/exec/format/parquet/vparquet_column_reader.cpp
b/be/src/vec/exec/format/parquet/vparquet_column_reader.cpp
index 75f43b4730..79760f82a6 100644
--- a/be/src/vec/exec/format/parquet/vparquet_column_reader.cpp
+++ b/be/src/vec/exec/format/parquet/vparquet_column_reader.cpp
@@ -27,7 +27,6 @@
namespace doris::vectorized {
Status ParquetColumnReader::create(FileReader* file, FieldSchema* field,
- const ParquetReadColumn& column,
const tparquet::RowGroup& row_group,
const std::vector<RowRange>& row_ranges,
cctz::time_zone* ctz,
std::unique_ptr<ParquetColumnReader>&
reader,
diff --git a/be/src/vec/exec/format/parquet/vparquet_column_reader.h
b/be/src/vec/exec/format/parquet/vparquet_column_reader.h
index de0ec185b9..f8d8085df8 100644
--- a/be/src/vec/exec/format/parquet/vparquet_column_reader.h
+++ b/be/src/vec/exec/format/parquet/vparquet_column_reader.h
@@ -107,7 +107,7 @@ public:
virtual Status read_column_data(ColumnPtr& doris_column, DataTypePtr& type,
ColumnSelectVector& select_vector, size_t
batch_size,
size_t* read_rows, bool* eof) = 0;
- static Status create(FileReader* file, FieldSchema* field, const
ParquetReadColumn& column,
+ static Status create(FileReader* file, FieldSchema* field,
const tparquet::RowGroup& row_group,
const std::vector<RowRange>& row_ranges,
cctz::time_zone* ctz,
std::unique_ptr<ParquetColumnReader>& reader, size_t
max_buf_size);
@@ -190,4 +190,4 @@ private:
level_t _EMPTY_ARRAY = -1;
level_t _NULL_ARRAY = -1;
};
-}; // namespace doris::vectorized
\ No newline at end of file
+}; // namespace doris::vectorized
diff --git a/be/src/vec/exec/format/parquet/vparquet_group_reader.cpp
b/be/src/vec/exec/format/parquet/vparquet_group_reader.cpp
index 24441896a9..e3f841f1b1 100644
--- a/be/src/vec/exec/format/parquet/vparquet_group_reader.cpp
+++ b/be/src/vec/exec/format/parquet/vparquet_group_reader.cpp
@@ -57,7 +57,7 @@ Status RowGroupReader::init(const FieldDescriptor& schema,
std::vector<RowRange>
for (auto& read_col : _read_columns) {
auto field =
const_cast<FieldSchema*>(schema.get_column(read_col._file_slot_name));
std::unique_ptr<ParquetColumnReader> reader;
- RETURN_IF_ERROR(ParquetColumnReader::create(_file_reader, field,
read_col, _row_group_meta,
+ RETURN_IF_ERROR(ParquetColumnReader::create(_file_reader, field,
_row_group_meta,
_read_ranges, _ctz,
reader, max_buf_size));
auto col_iter = col_offsets.find(read_col._parquet_col_id);
if (col_iter != col_offsets.end()) {
@@ -233,6 +233,7 @@ Status RowGroupReader::_do_lazy_read(Block* block, size_t
batch_size, size_t* re
DCHECK_EQ(pre_read_rows + _cached_filtered_rows, 0);
*read_rows = 0;
*batch_eof = true;
+ return Status::OK();
}
ColumnSelectVector& select_vector = *select_vector_ptr;
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@doris.apache.org
For additional commands, e-mail: commits-h...@doris.apache.org
