xiaokang commented on code in PR #38213:
URL: https://github.com/apache/doris/pull/38213#discussion_r1690951663
##########
be/src/vec/exec/format/json/new_json_reader.cpp:
##########
@@ -1657,7 +1657,19 @@ Status
NewJsonReader::_simdjson_write_columns_by_jsonpath(
return st;
}
}
- if (i >= _parsed_jsonpaths.size() || st.is<NOT_FOUND>()) {
+ if (i < _parsed_jsonpaths.size() &&
JsonFunctions::is_root_path(_parsed_jsonpaths[i])) {
+ // Indicate that the jsonpath is "$.", read the full root json
object, insert the original doc directly
Review Comment:
Does this feature, that load all fields in JSON to variant column, match the
need for users? I think usually users only need some fields in JSON, not all
fields.
##########
regression-test/data/load_p0/stream_load/test_json_load.out:
##########
@@ -250,3 +250,9 @@ test k2_value
-- !select29 --
10 \N
+
+-- !select30 --
+12345 {"k1":12345,"k2":"11111","k3":111111,"k4":[11111]}
{"k1":12345,"k2":"11111","k3":111111,"k4":[11111]} 111111
+12346 {"k1":12346,"k2":"22222","k4":[22222]}
{"k1":12346,"k2":"22222","k4":[22222]} \N
+12347 {"k1":12347,"k3":"33333","k5":{"k51":1024,"xxxx":[11111]}}
{"k1":12347,"k3":"33333","k5":{"k51":1024,"xxxx":[11111]}} 33333
+12347 {"k1":12347,"k3":"33333","k4":[22222]}
{"k1":12347,"k3":"33333","k4":[22222]} 33333
Review Comment:
unstable output since k1 is the same for row 3 and 4.
--
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]
