This is an automated email from the ASF dual-hosted git repository.
morningman pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/doris.git
The following commit(s) were added to refs/heads/master by this push:
new 675b5f17525 [fix](load)fix load json format boolean type to int
column. (#54397)
675b5f17525 is described below
commit 675b5f175253d412182612e37fcc07c699509674
Author: daidai <[email protected]>
AuthorDate: Wed Aug 13 06:51:04 2025 +0800
[fix](load)fix load json format boolean type to int column. (#54397)
### What problem does this PR solve?
Related PR: #43469
Problem Summary:
PR #43469 accidentally removed the logic for reading boolean
"true"/"false" values in the simd join reader. Before PR #43469,
"true"/"false" were treated as "1"/"0", allowing a bool column in a JSON
file to be imported into an Doris int column. This PR restores this
logic.
---
be/src/vec/exec/format/json/new_json_reader.cpp | 11 ++++++++
.../data/load_p0/stream_load/test_json_load.out | Bin 5107 -> 5185 bytes
.../stream_load/test_read_boolean_to_int.json | 6 +++++
.../load_p0/stream_load/test_json_load.groovy | 29 +++++++++++++++++++++
4 files changed, 46 insertions(+)
diff --git a/be/src/vec/exec/format/json/new_json_reader.cpp
b/be/src/vec/exec/format/json/new_json_reader.cpp
index fc2e23eb685..e49100f82a1 100644
--- a/be/src/vec/exec/format/json/new_json_reader.cpp
+++ b/be/src/vec/exec/format/json/new_json_reader.cpp
@@ -1057,6 +1057,17 @@ Status
NewJsonReader::_simdjson_write_data_to_column(simdjson::ondemand::value&
RETURN_IF_ERROR(data_serde->deserialize_one_cell_from_json(*data_column_ptr,
slice,
_serde_options));
+ } else if (value.type() == simdjson::ondemand::json_type::boolean) {
+ const char* str_value = nullptr;
+ // insert "1"/"0" , not "true"/"false".
+ if (value.get_bool()) {
+ str_value = (char*)"1";
+ } else {
+ str_value = (char*)"0";
+ }
+ Slice slice {str_value, 1};
+
RETURN_IF_ERROR(data_serde->deserialize_one_cell_from_json(*data_column_ptr,
slice,
+
_serde_options));
} else {
// Maybe we can `switch (value->GetType()) case: kNumberType`.
// Note that `if (value->IsInt())`, but column is FloatColumn.
diff --git a/regression-test/data/load_p0/stream_load/test_json_load.out
b/regression-test/data/load_p0/stream_load/test_json_load.out
index 3ef9ecb5be9..b2cf149e12b 100644
Binary files a/regression-test/data/load_p0/stream_load/test_json_load.out and
b/regression-test/data/load_p0/stream_load/test_json_load.out differ
diff --git
a/regression-test/data/load_p0/stream_load/test_read_boolean_to_int.json
b/regression-test/data/load_p0/stream_load/test_read_boolean_to_int.json
new file mode 100644
index 00000000000..62ff27be0bf
--- /dev/null
+++ b/regression-test/data/load_p0/stream_load/test_read_boolean_to_int.json
@@ -0,0 +1,6 @@
+[
+{"id":1,"k1":true,"k2":true,"k3":true,"k4":true},
+{"id":2,"k1":false,"k2":false,"k3":false,"k4":false},
+{"id":3,"k1":100,"k2":100,"k3":100,"k4":100},
+{"id":4,"k1":null,"k2":null,"k3":null,"k4":null}
+]
\ No newline at end of file
diff --git a/regression-test/suites/load_p0/stream_load/test_json_load.groovy
b/regression-test/suites/load_p0/stream_load/test_json_load.groovy
index 128d8fb6b0f..3f66745d941 100644
--- a/regression-test/suites/load_p0/stream_load/test_json_load.groovy
+++ b/regression-test/suites/load_p0/stream_load/test_json_load.groovy
@@ -943,4 +943,33 @@ suite("test_json_load", "p0,nonConcurrent") {
} finally {
// try_sql("DROP TABLE IF EXISTS ${testTable}")
}
+
+ // try to load `boolean` => `tinyint, int , string, decimal`
+ try {
+ sql "DROP TABLE IF EXISTS ${testTable}"
+ sql """CREATE TABLE IF NOT EXISTS ${testTable}
+ (
+ `id` int,
+ `k1` tinyint NULL,
+ `k2` int NULL,
+ `k3` string NULL,
+ `k4` decimal(10,2) NULL
+ )
+ DUPLICATE KEY(`id`)
+ COMMENT ''
+ DISTRIBUTED BY RANDOM BUCKETS 1
+ PROPERTIES (
+ "replication_allocation" = "tag.location.default: 1"
+ );"""
+
+
+ load_json_data.call("${testTable}", "${testTable}_case31", 'true', '',
'json', '', '',
+ '', '', '', 'test_read_boolean_to_int.json')
+
+ sql "sync"
+ qt_select31 "select * from ${testTable} order by id"
+
+ } finally {
+ // try_sql("DROP TABLE IF EXISTS ${testTable}")
+ }
}
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
