Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[fix](parquet) fix dict page offset zero incompatible behaviour #41506

Merged
merged 3 commits into from
Oct 10, 2024
Merged

[fix](parquet) fix dict page offset zero incompatible behaviour #41506

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
00bbb4e
fix dict page offset zero bug
suxiaogang223 Sep 30, 2024
215f451
fix
suxiaogang223 Oct 6, 2024
a1b18b1
extract _has_dict_page()
suxiaogang223 Oct 8, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 7 additions & 4 deletions be/src/vec/exec/format/parquet/vparquet_column_chunk_reader.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -59,16 +59,15 @@ ColumnChunkReader::ColumnChunkReader(io::BufferedStreamReader* reader,
_io_ctx(io_ctx) {}

Status ColumnChunkReader::init() {
size_t start_offset = _metadata.__isset.dictionary_page_offset
suxiaogang223 marked this conversation as resolved.
Show resolved Hide resolved
? _metadata.dictionary_page_offset
: _metadata.data_page_offset;
size_t start_offset =
_has_dict_page() ? _metadata.dictionary_page_offset : _metadata.data_page_offset;
size_t chunk_size = _metadata.total_compressed_size;
// create page reader
_page_reader = create_page_reader(_stream_reader, _io_ctx, start_offset, chunk_size,
_metadata.num_values, _offset_index);
// get the block compression codec
RETURN_IF_ERROR(get_block_compression_codec(_metadata.codec, &_block_compress_codec));
if (_metadata.__isset.dictionary_page_offset) {
if (_has_dict_page()) {
// seek to the directory page
_page_reader->seek_to_page(_metadata.dictionary_page_offset);
// Parse dictionary data when reading
Expand All @@ -82,6 +81,10 @@ Status ColumnChunkReader::init() {
return Status::OK();
}

bool ColumnChunkReader::_has_dict_page() const {
return _metadata.__isset.dictionary_page_offset && _metadata.dictionary_page_offset > 0;
}

Status ColumnChunkReader::next_page() {
if (_state == HEADER_PARSED) {
return Status::OK();
Expand Down
1 change: 1 addition & 0 deletions be/src/vec/exec/format/parquet/vparquet_column_chunk_reader.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -191,6 +191,7 @@ class ColumnChunkReader {
private:
enum ColumnChunkReaderState { NOT_INIT, INITIALIZED, HEADER_PARSED, DATA_LOADED, PAGE_SKIPPED };

bool _has_dict_page() const;
Status _decode_dict_page();
void _reserve_decompress_buf(size_t size);
int32_t _get_type_length();
Expand Down
7 changes: 4 additions & 3 deletions be/src/vec/exec/format/parquet/vparquet_column_reader.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -179,9 +179,10 @@ void ParquetColumnReader::_generate_read_ranges(int64_t start_index, int64_t end
Status ScalarColumnReader::init(io::FileReaderSPtr file, FieldSchema* field, size_t max_buf_size) {
_field_schema = field;
auto& chunk_meta = _chunk_meta.meta_data;
int64_t chunk_start = chunk_meta.__isset.dictionary_page_offset
? chunk_meta.dictionary_page_offset
: chunk_meta.data_page_offset;
int64_t chunk_start =
chunk_meta.__isset.dictionary_page_offset && chunk_meta.dictionary_page_offset > 0
? chunk_meta.dictionary_page_offset
: chunk_meta.data_page_offset;
size_t chunk_len = chunk_meta.total_compressed_size;
size_t prefetch_buffer_size = std::min(chunk_len, max_buf_size);
if (typeid_cast<io::MergeRangeFileReader*>(file.get())) {
Expand Down
Binary file modified regression-test/data/external_table_p0/tvf/test_hdfs_parquet_group0.out
View file
Open in desktop
Binary file not shown.
5 changes: 1 addition & 4 deletions regression-test/suites/external_table_p0/tvf/test_hdfs_parquet_group0.groovy
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -223,13 +223,10 @@ suite("test_hdfs_parquet_group0","external,hive,tvf,external_docker") {


uri = "${defaultFS}" + "/user/doris/tvf_data/test_hdfs_parquet/group0/dict-page-offset-zero.parquet"
test {
sql """ select * from HDFS(
order_qt_test_27 """ select * from HDFS(
"uri" = "${uri}",
"hadoop.username" = "${hdfsUserName}",
"format" = "parquet") limit 10; """
exception "Failed to deserialize parquet page header. offset: 0, header size: 40, end offset: 40, real header size: 40"
}


uri = "${defaultFS}" + "/user/doris/tvf_data/test_hdfs_parquet/group0/repeated_no_annotation.parquet"
Expand Down
11 changes: 0 additions & 11 deletions regression-test/suites/external_table_p0/tvf/test_hdfs_parquet_group5.groovy
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -218,17 +218,6 @@ suite("test_hdfs_parquet_group5","external,hive,tvf,external_docker") {
"hadoop.username" = "${hdfsUserName}",
"format" = "parquet") limit 10; """


uri = "${defaultFS}" + "/user/doris/tvf_data/test_hdfs_parquet/group5/dict-page-offset-zero.parquet"
test {
sql """ select * from HDFS(
"uri" = "${uri}",
"hadoop.username" = "${hdfsUserName}",
"format" = "parquet") limit 10; """
exception "Failed to deserialize parquet page header. offset: 0, header size: 40, end offset: 40, real header size: 40"
}


uri = "${defaultFS}" + "/user/doris/tvf_data/test_hdfs_parquet/group5/native_parquet_reader.parquet"
order_qt_test_28 """ select * from HDFS(
"uri" = "${uri}",
Expand Down
Loading