You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@doris.apache.org by GitBox <gi...@apache.org> on 2022/08/15 11:48:54 UTC
[GitHub] [doris] wsjz commented on a diff in pull request #11601: [feature-wip](parquet-reader) update column read model and add page index
wsjz commented on code in PR #11601:
URL: https://github.com/apache/doris/pull/11601#discussion_r945651384
##########
be/src/vec/exec/format/parquet/vparquet_reader.cpp:
##########
@@ -90,63 +95,240 @@ Status ParquetReader::_init_read_columns(const std::vector<SlotDescriptor*>& tup
auto physical_type = _file_metadata->schema().get_column(parquet_col_id)->physical_type;
column.parquet_type = physical_type;
_read_columns.emplace_back(column);
+ VLOG_DEBUG << "slot_desc " << slot_desc->debug_string();
}
return Status::OK();
}
-Status ParquetReader::read_next_batch(Block* block) {
- int32_t group_id = 0;
- RETURN_IF_ERROR(_row_group_reader->get_next_row_group(&group_id));
+Status ParquetReader::read_next_batch(Block* block, bool* eof) {
+ DCHECK(_total_groups == _row_group_readers.size());
+ if (_total_groups == 0) {
+ *eof = true;
+ }
+ bool _batch_eof = false;
+ auto row_group_reader = _row_group_readers[_current_row_group_id];
+ RETURN_IF_ERROR(row_group_reader->next_batch(block, _batch_size, &_batch_eof));
+ if (_batch_eof) {
+ _current_row_group_id++;
+ if (_current_row_group_id > _total_groups) {
+ *eof = true;
+ }
Review Comment:
The `_current_row_group_id++;` statement can do this，so it's implicit to mark `*eof = true`
##########
be/src/vec/exec/format/parquet/vparquet_reader.cpp:
##########
@@ -90,63 +95,240 @@ Status ParquetReader::_init_read_columns(const std::vector<SlotDescriptor*>& tup
auto physical_type = _file_metadata->schema().get_column(parquet_col_id)->physical_type;
column.parquet_type = physical_type;
_read_columns.emplace_back(column);
+ VLOG_DEBUG << "slot_desc " << slot_desc->debug_string();
}
return Status::OK();
}
-Status ParquetReader::read_next_batch(Block* block) {
- int32_t group_id = 0;
- RETURN_IF_ERROR(_row_group_reader->get_next_row_group(&group_id));
+Status ParquetReader::read_next_batch(Block* block, bool* eof) {
+ DCHECK(_total_groups == _row_group_readers.size());
+ if (_total_groups == 0) {
+ *eof = true;
+ }
+ bool _batch_eof = false;
+ auto row_group_reader = _row_group_readers[_current_row_group_id];
+ RETURN_IF_ERROR(row_group_reader->next_batch(block, _batch_size, &_batch_eof));
+ if (_batch_eof) {
+ _current_row_group_id++;
+ if (_current_row_group_id > _total_groups) {
+ *eof = true;
+ }
Review Comment:
The `_current_row_group_id++;` statement can do this，so it's implicit to mark `*eof =false`
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@doris.apache.org
For additional commands, e-mail: commits-help@doris.apache.org