INSERT ROW part of MERGE statement dont seem to be supported

[sbv-csis]

When using the INSERT ROW part of merge it seems to fail - when trying to use https://cloud.google.com/bigquery/docs/reference/standard-sql/dml-syntax#omitting_column_names_2

bq --api http://127.0.0.1:9050 query "MERGE test-dataset-bla.t1 t using (select * from test-dataset-bla.t2) s on s.name = t.name when not matched by target then insert row when not matched by source then delete"
BigQuery error in query operation: Error processing job 'test-project-bla:bqjob_r554b87e1dbc3244d_00000186c0a89d32_1': failed to analyze:
INVALID_ARGUMENT: Missing insert column list [at 1:124]

as far as I can read zetasql should support it - though with some kind of feature flag FEATURE_V_1_3_OMIT_INSERT_COLUMN_LIST

And when trying with explict select and values lists it still fails:

E0308 11:06:03.384493 140062286534464 bq_utils.py:245] BigQuery error in query operation: Error processing job 'test-project-bla:bqjob_r33ab615fd2153cbd_00000186c0af5c11_1': failed to exec merge
statement CREATE TABLE zetasqlite_merged_table AS SELECT DISTINCT * FROM (SELECT * FROM SELECT `name#2` FROM (SELECT `name` AS `name#2` FROM
`test-project-bla_test-dataset-bla_t2`) LEFT JOIN (SELECT `name` AS `name#1` FROM `test-project-bla_test-dataset-bla_t1`) ON
zetasqlite_equal(`name#2`,`name#1`) UNION ALL SELECT * FROM (SELECT `name` AS `name#1` FROM `test-project-bla_test-dataset-bla_t1`) LEFT
JOIN SELECT `name#2` FROM (SELECT `name` AS `name#2` FROM `test-project-bla_test-dataset-bla_t2`) ON zetasqlite_equal(`name#2`,`name#1`)):
near "SELECT": syntax error
Traceback (most recent call last):
  File "/home/sbv/google-cloud-sdk/platform/bq/bq.py", line 856, in RunSafely
    return_value = self.RunWithArgs(*args, **kwds)
  File "/home/sbv/google-cloud-sdk/platform/bq/bq.py", line 2144, in RunWithArgs
    job = client.Query(query, **kwds)
  File "/home/sbv/google-cloud-sdk/platform/bq/bigquery_client.py", line 6563, in Query
    return self.ExecuteJob(request, **kwds)
  File "/home/sbv/google-cloud-sdk/platform/bq/bigquery_client.py", line 6046, in ExecuteJob
    job = self.RunJobSynchronously(
  File "/home/sbv/google-cloud-sdk/platform/bq/bigquery_client.py", line 6032, in RunJobSynchronously
    return self.RaiseIfJobError(result)
  File "/home/sbv/google-cloud-sdk/platform/bq/bigquery_client.py", line 3286, in RaiseIfJobError
    raise BigqueryError.Create(
bigquery_client.BigqueryServiceError: Error processing job 'test-project-bla:bqjob_r33ab615fd2153cbd_00000186c0af5c11_1': failed to exec merge statement CREATE TABLE zetasqlite_merged_table AS SELECT DISTINCT * FROM (SELECT * FROM SELECT `name#2` FROM (SELECT `name` AS `name#2` FROM `test-project-bla_test-dataset-bla_t2`) LEFT JOIN (SELECT `name` AS `name#1` FROM `test-project-bla_test-dataset-bla_t1`) ON zetasqlite_equal(`name#2`,`name#1`) UNION ALL SELECT * FROM (SELECT `name` AS `name#1` FROM `test-project-bla_test-dataset-bla_t1`) LEFT JOIN SELECT `name#2` FROM (SELECT `name` AS `name#2` FROM `test-project-bla_test-dataset-bla_t2`) ON zetasqlite_equal(`name#2`,`name#1`)): near "SELECT": syntax error
BigQuery error in query operation: Error processing job 'test-project-bla:bqjob_r33ab615fd2153cbd_00000186c0af5c11_1': failed to exec merge
statement CREATE TABLE zetasqlite_merged_table AS SELECT DISTINCT * FROM (SELECT * FROM SELECT `name#2` FROM (SELECT `name` AS `name#2` FROM
`test-project-bla_test-dataset-bla_t2`) LEFT JOIN (SELECT `name` AS `name#1` FROM `test-project-bla_test-dataset-bla_t1`) ON
zetasqlite_equal(`name#2`,`name#1`) UNION ALL SELECT * FROM (SELECT `name` AS `name#1` FROM `test-project-bla_test-dataset-bla_t1`) LEFT
JOIN SELECT `name#2` FROM (SELECT `name` AS `name#2` FROM `test-project-bla_test-dataset-bla_t2`) ON zetasqlite_equal(`name#2`,`name#1`)):
near "SELECT": syntax error

[sbv-csis]

And t1 and t2 are simple one column tables:

$ bq --api http://127.0.0.1:9050 --project_id=test-project-bla query "select * from test-dataset-bla.t2";
+------+
| name |
+------+
| n1   |
+------+
$  bq --api http://127.0.0.1:9050 --project_id=test-project-bla query "select * from test-dataset-bla.t1";

$

[bony2023]

This should be fixed by this PR: https://github.com/goccy/go-zetasqlite/pull/222 cc - @goccy