github-actions[bot] commented on code in PR #69591:
URL: https://github.com/apache/doris/pull/69591#discussion_r4373928157
##########
regression-test/suites/query_p0/test_array_with_single_row_max_than_uint16.groovy:
##########
@@ -51,11 +36,13 @@ suite("test_array_with_single_row_max_than_uint16",
"query") {
)
"""
// prepare data
- qt_sql """
- insert into ${testTable} select * from local(
- "file_path" = "${outFilePath}/arr_max.orc",
- "backend_id" = "${be_id}",
- "format" = "orc");"""
+ streamLoad {
+ table "${testTable}"
+ set 'format', 'orc'
Review Comment:
[P2] Use a load path that supports these complex ORC columns
This table has the ARRAY column `c_decimal`, and the same migration in
`regression-test/suites/types/complex_types/map_uniq_with_local_tvf.groovy:37-41`
targets a MAP column. ORC Stream Load uses the legacy `FileScanner`:
`FileScanLocalState::should_use_file_scanner_v2()` explicitly excludes
`is_load`, and `_init_src_block_for_load()` calls `_check_output_block_types()`
(`be/src/exec/scan/file_scanner.cpp:628-638`), which rejects either destination
with `Parquet/orc doesn't support complex types in broker/stream load`. The
default `streamLoad` checker throws on that failure, so both migrated suites
stop before their existing count/value assertions. Please load these fixtures
through a supported remote TVF or another format/path that supports complex
columns, preserving the existing assertions.
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
