yuanlihan commented on a change in pull request #1629: Enable parsing columns
from file path for Broker Load (#1582)
URL: https://github.com/apache/incubator-doris/pull/1629#discussion_r313248654
##########
File path: be/src/exec/parquet_reader.cpp
##########
@@ -204,13 +208,23 @@ Status ParquetReaderWrap::read(Tuple* tuple, const
std::vector<SlotDescriptor*>&
uint8_t tmp_buf[128] = {0};
int32_t wbytes = 0;
const uint8_t *value = nullptr;
+ int index = 0;
int column_index = 0;
try {
size_t slots = tuple_slot_descs.size();
for (size_t i = 0; i < slots; ++i) {
auto slot_desc = tuple_slot_descs[i];
- column_index = i;// column index in batch record
- switch (_parquet_column_type[i]) {
+ auto iter = _columns_from_path.find(slot_desc->col_name());
Review comment:
Currently, we have two broker scanners. One is BrokerScanner using for
reading CSV like row-storage sources. The other is ParquetScanner using for
reading parquet files, which is a column-storage source. It seems that we
cannot reuse the `line_to_src_tuple` method of BrokerScanner
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@doris.apache.org
For additional commands, e-mail: dev-h...@doris.apache.org
