You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@spark.apache.org by we...@apache.org on 2026/03/16 07:38:33 UTC
(spark) branch master updated: [SPARK-55984][SQL][TESTS] Add metadata_column_resolution.sql golden file
This is an automated email from the ASF dual-hosted git repository.
wenchen pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/spark.git
The following commit(s) were added to refs/heads/master by this push:
new ceba3da57491 [SPARK-55984][SQL][TESTS] Add metadata_column_resolution.sql golden file
ceba3da57491 is described below
commit ceba3da5749197497a6690746621bae3df1be5d7
Author: mihailoale-db <mi...@databricks.com>
AuthorDate: Mon Mar 16 15:38:15 2026 +0800
[SPARK-55984][SQL][TESTS] Add metadata_column_resolution.sql golden file
### What changes were proposed in this pull request?
In this issue I propose to test `_metadata` column resolution by adding `metadata_column_resolution.sql` golden file.
### Why are the changes needed?
Extending the tests.
### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?
Test only change.
### Was this patch authored or co-authored using generative AI tooling?
No.
Closes #54783 from mihailoale-db/addmetadatatests.
Authored-by: mihailoale-db <mi...@databricks.com>
Signed-off-by: Wenchen Fan <we...@databricks.com>
---
.../metadata_column_resolution.sql.out | 542 +++++++++++++++++++++
.../inputs/metadata_column_resolution.sql | 129 +++++
.../results/metadata_column_resolution.sql.out | 493 +++++++++++++++++++
3 files changed, 1164 insertions(+)
diff --git a/sql/core/src/test/resources/sql-tests/analyzer-results/metadata_column_resolution.sql.out b/sql/core/src/test/resources/sql-tests/analyzer-results/metadata_column_resolution.sql.out
new file mode 100644
index 000000000000..a4635a58fc6e
--- /dev/null
+++ b/sql/core/src/test/resources/sql-tests/analyzer-results/metadata_column_resolution.sql.out
@@ -0,0 +1,542 @@
+-- Automatically generated by SQLQueryTestSuite
+-- !query
+CREATE DATABASE IF NOT EXISTS testdb
+-- !query analysis
+CreateNamespace true
++- ResolvedNamespace V2SessionCatalog(spark_catalog), [testdb]
+
+
+-- !query
+USE testdb
+-- !query analysis
+SetCatalogAndNamespace
++- ResolvedNamespace V2SessionCatalog(spark_catalog), [testdb]
+
+
+-- !query
+CREATE TABLE t1(id INT, name STRING, value DOUBLE) USING PARQUET
+-- !query analysis
+CreateDataSourceTableCommand `spark_catalog`.`testdb`.`t1`, false
+
+
+-- !query
+CREATE TABLE t2(id INT, category STRING, amount BIGINT) USING CSV
+-- !query analysis
+CreateDataSourceTableCommand `spark_catalog`.`testdb`.`t2`, false
+
+
+-- !query
+CREATE TABLE t3(a INT, b STRING, c DOUBLE) USING ORC
+-- !query analysis
+CreateDataSourceTableCommand `spark_catalog`.`testdb`.`t3`, false
+
+
+-- !query
+SELECT _metadata.file_name FROM t1
+-- !query analysis
+Project [_metadata#x.file_name AS file_name#x]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT _metadata.file_path FROM t1
+-- !query analysis
+Project [_metadata#x.file_path AS file_path#x]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT _metadata.file_size FROM t1
+-- !query analysis
+Project [_metadata#x.file_size AS file_size#xL]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT _metadata.file_modification_time FROM t1
+-- !query analysis
+Project [_metadata#x.file_modification_time AS file_modification_time#x]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT _metadata.row_index FROM t1
+-- !query analysis
+Project [_metadata#x.row_index AS row_index#xL]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT _metadata.file_block_start FROM t1
+-- !query analysis
+Project [_metadata#x.file_block_start AS file_block_start#xL]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT _metadata.file_block_length FROM t1
+-- !query analysis
+Project [_metadata#x.file_block_length AS file_block_length#xL]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT t1._metadata.file_name FROM t1
+-- !query analysis
+Project [_metadata#x.file_name AS file_name#x]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT testdb.t1._metadata.file_name FROM testdb.t1
+-- !query analysis
+Project [_metadata#x.file_name AS file_name#x]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT _metadata.file_name, t1._metadata.file_path FROM t1
+-- !query analysis
+Project [_metadata#x.file_name AS file_name#x, _metadata#x.file_path AS file_path#x]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT t1._metadata.file_name, testdb.t1._metadata.file_path FROM testdb.t1
+-- !query analysis
+Project [_metadata#x.file_name AS file_name#x, _metadata#x.file_path AS file_path#x]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT * FROM t1 WHERE _metadata.file_size > 1000
+-- !query analysis
+Project [id#x, name#x, value#x]
++- Project [id#x, name#x, value#x]
+ +- Filter (_metadata#x.file_size > cast(1000 as bigint))
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT id, name FROM t1 WHERE _metadata.row_index < 100
+-- !query analysis
+Project [id#x, name#x]
++- Project [id#x, name#x, value#x]
+ +- Filter (_metadata#x.row_index < cast(100 as bigint))
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT * FROM t1 WHERE _metadata.file_name = 'data.parquet'
+-- !query analysis
+Project [id#x, name#x, value#x]
++- Project [id#x, name#x, value#x]
+ +- Filter (_metadata#x.file_name = data.parquet)
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT * FROM t1 WHERE _metadata.file_modification_time > timestamp'2024-01-01 00:00:00'
+-- !query analysis
+[Analyzer test output redacted due to nondeterminism]
+
+
+-- !query
+SELECT * FROM t1 WHERE t1._metadata.file_size > 1000
+-- !query analysis
+Project [id#x, name#x, value#x]
++- Project [id#x, name#x, value#x]
+ +- Filter (_metadata#x.file_size > cast(1000 as bigint))
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT * FROM testdb.t1 WHERE testdb.t1._metadata.row_index < 100
+-- !query analysis
+Project [id#x, name#x, value#x]
++- Project [id#x, name#x, value#x]
+ +- Filter (_metadata#x.row_index < cast(100 as bigint))
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT id, name, _metadata.file_name FROM t1 ORDER BY _metadata.file_size
+-- !query analysis
+Project [id#x, name#x, file_name#x]
++- Sort [_metadata#x.file_size ASC NULLS FIRST], true
+ +- Project [id#x, name#x, _metadata#x.file_name AS file_name#x, _metadata#x]
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT * FROM t1 ORDER BY _metadata.row_index DESC
+-- !query analysis
+Project [id#x, name#x, value#x]
++- Sort [_metadata#x.row_index DESC NULLS LAST], true
+ +- Project [id#x, name#x, value#x, _metadata#x]
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT id, _metadata.file_name FROM t1 ORDER BY _metadata.file_modification_time, _metadata.row_index
+-- !query analysis
+Project [id#x, file_name#x]
++- Sort [_metadata#x.file_modification_time ASC NULLS FIRST, _metadata#x.row_index ASC NULLS FIRST], true
+ +- Project [id#x, _metadata#x.file_name AS file_name#x, _metadata#x]
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT t1._metadata.file_name, COUNT(*) FROM t1 GROUP BY t1._metadata.file_name
+-- !query analysis
+Aggregate [_metadata#x.file_name], [_metadata#x.file_name AS file_name#x, count(1) AS count(1)#xL]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT testdb.t1._metadata.file_path, SUM(value) FROM testdb.t1 GROUP BY testdb.t1._metadata.file_path
+-- !query analysis
+Aggregate [_metadata#x.file_path], [_metadata#x.file_path AS file_path#x, sum(value#x) AS sum(value)#x]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT _metadata.file_name, MIN(_metadata.row_index), MAX(_metadata.row_index) FROM t1
+GROUP BY _metadata.file_name
+-- !query analysis
+Aggregate [_metadata#x.file_name], [_metadata#x.file_name AS file_name#x, min(_metadata#x.row_index) AS min(_metadata.row_index)#xL, max(_metadata#x.row_index) AS max(_metadata.row_index)#xL]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT _metadata.file_path, COUNT(DISTINCT _metadata.file_name) FROM t1
+GROUP BY _metadata.file_path
+-- !query analysis
+Aggregate [_metadata#x.file_path], [_metadata#x.file_path AS file_path#x, count(distinct _metadata#x.file_name) AS count(DISTINCT _metadata.file_name)#xL]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT t1.id, t2.category, t1._metadata.file_name
+FROM t1 JOIN t2 ON t1.id = t2.id
+-- !query analysis
+Project [id#x, category#x, _metadata#x.file_name AS file_name#x]
++- Join Inner, (id#x = id#x)
+ :- SubqueryAlias spark_catalog.testdb.t1
+ : +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+ +- SubqueryAlias spark_catalog.testdb.t2
+ +- Relation spark_catalog.testdb.t2[id#x,category#x,amount#xL] csv
+
+
+-- !query
+SELECT t1.id, t1._metadata.file_name, t2._metadata.file_name
+FROM t1 JOIN t2 ON t1.id = t2.id
+-- !query analysis
+Project [id#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_name AS file_name#x]
++- Join Inner, (id#x = id#x)
+ :- SubqueryAlias spark_catalog.testdb.t1
+ : +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+ +- SubqueryAlias spark_catalog.testdb.t2
+ +- Relation spark_catalog.testdb.t2[id#x,category#x,amount#xL,_metadata#x] csv
+
+
+-- !query
+SELECT t1.id, t1._metadata.row_index, t2._metadata.row_index
+FROM t1 JOIN t2 ON t1.id = t2.id
+WHERE t1._metadata.file_size > 1000
+-- !query analysis
+org.apache.spark.sql.AnalysisException
+{
+ "errorClass" : "FIELD_NOT_FOUND",
+ "sqlState" : "42704",
+ "messageParameters" : {
+ "fieldName" : "`row_index`",
+ "fields" : "`file_path`, `file_name`, `file_size`, `file_block_start`, `file_block_length`, `file_modification_time`"
+ },
+ "queryContext" : [ {
+ "objectType" : "",
+ "objectName" : "",
+ "startIndex" : 39,
+ "stopIndex" : 60,
+ "fragment" : "t2._metadata.row_index"
+ } ]
+}
+
+
+-- !query
+SELECT t1.name, t1._metadata.file_name, t2._metadata.file_name
+FROM t1 LEFT JOIN t2 ON t1.id = t2.id
+-- !query analysis
+Project [name#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_name AS file_name#x]
++- Join LeftOuter, (id#x = id#x)
+ :- SubqueryAlias spark_catalog.testdb.t1
+ : +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+ +- SubqueryAlias spark_catalog.testdb.t2
+ +- Relation spark_catalog.testdb.t2[id#x,category#x,amount#xL,_metadata#x] csv
+
+
+-- !query
+SELECT t1._metadata.file_path, t2._metadata.file_path, t1.id
+FROM t1 RIGHT JOIN t2 ON t1.id = t2.id
+-- !query analysis
+Project [_metadata#x.file_path AS file_path#x, _metadata#x.file_path AS file_path#x, id#x]
++- Join RightOuter, (id#x = id#x)
+ :- SubqueryAlias spark_catalog.testdb.t1
+ : +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+ +- SubqueryAlias spark_catalog.testdb.t2
+ +- Relation spark_catalog.testdb.t2[id#x,category#x,amount#xL,_metadata#x] csv
+
+
+-- !query
+SELECT t1._metadata.file_name, t2._metadata.file_name, t1.id, t2.id
+FROM t1 FULL OUTER JOIN t2 ON t1.id = t2.id
+-- !query analysis
+Project [_metadata#x.file_name AS file_name#x, _metadata#x.file_name AS file_name#x, id#x, id#x]
++- Join FullOuter, (id#x = id#x)
+ :- SubqueryAlias spark_catalog.testdb.t1
+ : +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+ +- SubqueryAlias spark_catalog.testdb.t2
+ +- Relation spark_catalog.testdb.t2[id#x,category#x,amount#xL,_metadata#x] csv
+
+
+-- !query
+SELECT a.id, a._metadata.file_name, b._metadata.file_name
+FROM t1 a JOIN t1 b ON a.id = b.id
+WHERE a._metadata.row_index < b._metadata.row_index
+-- !query analysis
+Project [id#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_name AS file_name#x]
++- Filter (_metadata#x.row_index < _metadata#x.row_index)
+ +- Join Inner, (id#x = id#x)
+ :- SubqueryAlias a
+ : +- SubqueryAlias spark_catalog.testdb.t1
+ : +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+ +- SubqueryAlias b
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT t._metadata.file_name, t.id FROM t1 t
+-- !query analysis
+Project [_metadata#x.file_name AS file_name#x, id#x]
++- SubqueryAlias t
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT x._metadata.file_name, x._metadata.row_index FROM t1 x
+WHERE x._metadata.file_size > 500
+-- !query analysis
+Project [_metadata#x.file_name AS file_name#x, _metadata#x.row_index AS row_index#xL]
++- Project [id#x, name#x, value#x, _metadata#x]
+ +- Filter (_metadata#x.file_size > cast(500 as bigint))
+ +- SubqueryAlias x
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT * FROM (SELECT id, _metadata.file_name FROM t1) sub
+-- !query analysis
+Project [id#x, file_name#x]
++- SubqueryAlias sub
+ +- Project [id#x, _metadata#x.file_name AS file_name#x]
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT id FROM t1 WHERE _metadata.file_name IN
+ (SELECT _metadata.file_name FROM t2 WHERE amount > 100)
+-- !query analysis
+Project [id#x]
++- Project [id#x, name#x, value#x]
+ +- Filter _metadata#x.file_name IN (list#x [])
+ : +- Project [_metadata#x.file_name AS file_name#x]
+ : +- Filter (amount#xL > cast(100 as bigint))
+ : +- SubqueryAlias spark_catalog.testdb.t2
+ : +- Relation spark_catalog.testdb.t2[id#x,category#x,amount#xL,_metadata#x] csv
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT _metadata.file_name, id FROM t1
+UNION
+SELECT _metadata.file_name, id FROM t2
+-- !query analysis
+Distinct
++- Union false, false
+ :- Project [_metadata#x.file_name AS file_name#x, id#x]
+ : +- SubqueryAlias spark_catalog.testdb.t1
+ : +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+ +- Project [_metadata#x.file_name AS file_name#x, id#x]
+ +- SubqueryAlias spark_catalog.testdb.t2
+ +- Relation spark_catalog.testdb.t2[id#x,category#x,amount#xL,_metadata#x] csv
+
+
+-- !query
+SELECT _metadata.* FROM t1
+-- !query analysis
+Project [_metadata#x.file_path AS file_path#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_size AS file_size#xL, _metadata#x.file_block_start AS file_block_start#xL, _metadata#x.file_block_length AS file_block_length#xL, _metadata#x.file_modification_time AS file_modification_time#x, _metadata#x.row_index AS row_index#xL]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT _metadata.* FROM t2
+-- !query analysis
+Project [_metadata#x.file_path AS file_path#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_size AS file_size#xL, _metadata#x.file_block_start AS file_block_start#xL, _metadata#x.file_block_length AS file_block_length#xL, _metadata#x.file_modification_time AS file_modification_time#x]
++- SubqueryAlias spark_catalog.testdb.t2
+ +- Relation spark_catalog.testdb.t2[id#x,category#x,amount#xL,_metadata#x] csv
+
+
+-- !query
+SELECT _metadata.* FROM t3
+-- !query analysis
+Project [_metadata#x.file_path AS file_path#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_size AS file_size#xL, _metadata#x.file_block_start AS file_block_start#xL, _metadata#x.file_block_length AS file_block_length#xL, _metadata#x.file_modification_time AS file_modification_time#x]
++- SubqueryAlias spark_catalog.testdb.t3
+ +- Relation spark_catalog.testdb.t3[a#x,b#x,c#x,_metadata#x] orc
+
+
+-- !query
+SELECT t1._metadata.* FROM t1
+-- !query analysis
+Project [_metadata#x.file_path AS file_path#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_size AS file_size#xL, _metadata#x.file_block_start AS file_block_start#xL, _metadata#x.file_block_length AS file_block_length#xL, _metadata#x.file_modification_time AS file_modification_time#x, _metadata#x.row_index AS row_index#xL]
++- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT t2._metadata.* FROM t2
+-- !query analysis
+Project [_metadata#x.file_path AS file_path#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_size AS file_size#xL, _metadata#x.file_block_start AS file_block_start#xL, _metadata#x.file_block_length AS file_block_length#xL, _metadata#x.file_modification_time AS file_modification_time#x]
++- SubqueryAlias spark_catalog.testdb.t2
+ +- Relation spark_catalog.testdb.t2[id#x,category#x,amount#xL,_metadata#x] csv
+
+
+-- !query
+SELECT t3._metadata.* FROM t3
+-- !query analysis
+Project [_metadata#x.file_path AS file_path#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_size AS file_size#xL, _metadata#x.file_block_start AS file_block_start#xL, _metadata#x.file_block_length AS file_block_length#xL, _metadata#x.file_modification_time AS file_modification_time#x]
++- SubqueryAlias spark_catalog.testdb.t3
+ +- Relation spark_catalog.testdb.t3[a#x,b#x,c#x,_metadata#x] orc
+
+
+-- !query
+SELECT id, name, _metadata.* FROM t1 ORDER BY id
+-- !query analysis
+Sort [id#x ASC NULLS FIRST], true
++- Project [id#x, name#x, _metadata#x.file_path AS file_path#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_size AS file_size#xL, _metadata#x.file_block_start AS file_block_start#xL, _metadata#x.file_block_length AS file_block_length#xL, _metadata#x.file_modification_time AS file_modification_time#x, _metadata#x.row_index AS row_index#xL]
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT id, category, _metadata.* FROM t2 ORDER BY id
+-- !query analysis
+Sort [id#x ASC NULLS FIRST], true
++- Project [id#x, category#x, _metadata#x.file_path AS file_path#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_size AS file_size#xL, _metadata#x.file_block_start AS file_block_start#xL, _metadata#x.file_block_length AS file_block_length#xL, _metadata#x.file_modification_time AS file_modification_time#x]
+ +- SubqueryAlias spark_catalog.testdb.t2
+ +- Relation spark_catalog.testdb.t2[id#x,category#x,amount#xL,_metadata#x] csv
+
+
+-- !query
+SELECT sub.file_name, sub.file_size FROM (SELECT _metadata.* FROM t1) sub ORDER BY sub.file_name
+-- !query analysis
+Sort [file_name#x ASC NULLS FIRST], true
++- Project [file_name#x, file_size#xL]
+ +- SubqueryAlias sub
+ +- Project [_metadata#x.file_path AS file_path#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_size AS file_size#xL, _metadata#x.file_block_start AS file_block_start#xL, _metadata#x.file_block_length AS file_block_length#xL, _metadata#x.file_modification_time AS file_modification_time#x, _metadata#x.row_index AS row_index#xL]
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT id, _metadata.* FROM t1 WHERE _metadata.file_size >= 0 ORDER BY id
+-- !query analysis
+Sort [id#x ASC NULLS FIRST], true
++- Project [id#x, _metadata#x.file_path AS file_path#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_size AS file_size#xL, _metadata#x.file_block_start AS file_block_start#xL, _metadata#x.file_block_length AS file_block_length#xL, _metadata#x.file_modification_time AS file_modification_time#x, _metadata#x.row_index AS row_index#xL]
+ +- Project [id#x, name#x, value#x, _metadata#x]
+ +- Filter (_metadata#x.file_size >= cast(0 as bigint))
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT _metadata.*, COUNT(*) AS cnt FROM t1 GROUP BY ALL ORDER BY _metadata.file_name
+-- !query analysis
+Sort [file_name#x ASC NULLS FIRST], true
++- Aggregate [_metadata#x.file_path, _metadata#x.file_name, _metadata#x.file_size, _metadata#x.file_block_start, _metadata#x.file_block_length, _metadata#x.file_modification_time, _metadata#x.row_index], [_metadata#x.file_path AS file_path#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_size AS file_size#xL, _metadata#x.file_block_start AS file_block_start#xL, _metadata#x.file_block_length AS file_block_length#xL, _metadata#x.file_modification_time AS file_modification_time#x, [...]
+ +- SubqueryAlias spark_catalog.testdb.t1
+ +- Relation spark_catalog.testdb.t1[id#x,name#x,value#x,_metadata#x] parquet
+
+
+-- !query
+SELECT _metadata.*, COUNT(*) AS cnt FROM t2 GROUP BY ALL ORDER BY _metadata.file_name
+-- !query analysis
+Sort [file_name#x ASC NULLS FIRST], true
++- Aggregate [_metadata#x.file_path, _metadata#x.file_name, _metadata#x.file_size, _metadata#x.file_block_start, _metadata#x.file_block_length, _metadata#x.file_modification_time], [_metadata#x.file_path AS file_path#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_size AS file_size#xL, _metadata#x.file_block_start AS file_block_start#xL, _metadata#x.file_block_length AS file_block_length#xL, _metadata#x.file_modification_time AS file_modification_time#x, count(1) AS cnt#xL]
+ +- SubqueryAlias spark_catalog.testdb.t2
+ +- Relation spark_catalog.testdb.t2[id#x,category#x,amount#xL,_metadata#x] csv
+
+
+-- !query
+SELECT _metadata.*, COUNT(*) AS cnt FROM t3 GROUP BY ALL ORDER BY _metadata.file_name
+-- !query analysis
+Sort [file_name#x ASC NULLS FIRST], true
++- Aggregate [_metadata#x.file_path, _metadata#x.file_name, _metadata#x.file_size, _metadata#x.file_block_start, _metadata#x.file_block_length, _metadata#x.file_modification_time], [_metadata#x.file_path AS file_path#x, _metadata#x.file_name AS file_name#x, _metadata#x.file_size AS file_size#xL, _metadata#x.file_block_start AS file_block_start#xL, _metadata#x.file_block_length AS file_block_length#xL, _metadata#x.file_modification_time AS file_modification_time#x, count(1) AS cnt#xL]
+ +- SubqueryAlias spark_catalog.testdb.t3
+ +- Relation spark_catalog.testdb.t3[a#x,b#x,c#x,_metadata#x] orc
+
+
+-- !query
+DROP TABLE t1
+-- !query analysis
+DropTable false, false
++- ResolvedIdentifier V2SessionCatalog(spark_catalog), testdb.t1
+
+
+-- !query
+DROP TABLE t2
+-- !query analysis
+DropTable false, false
++- ResolvedIdentifier V2SessionCatalog(spark_catalog), testdb.t2
+
+
+-- !query
+DROP TABLE t3
+-- !query analysis
+DropTable false, false
++- ResolvedIdentifier V2SessionCatalog(spark_catalog), testdb.t3
+
+
+-- !query
+DROP DATABASE testdb
+-- !query analysis
+DropNamespace false, false
++- ResolvedNamespace V2SessionCatalog(spark_catalog), [testdb]
diff --git a/sql/core/src/test/resources/sql-tests/inputs/metadata_column_resolution.sql b/sql/core/src/test/resources/sql-tests/inputs/metadata_column_resolution.sql
new file mode 100644
index 000000000000..2a77a3abc415
--- /dev/null
+++ b/sql/core/src/test/resources/sql-tests/inputs/metadata_column_resolution.sql
@@ -0,0 +1,129 @@
+-- Tests for _metadata virtual column resolution across different file formats,
+-- qualifications, clauses, and join types.
+-- t1 uses PARQUET, t2 uses CSV, t3 uses ORC.
+
+CREATE DATABASE IF NOT EXISTS testdb;
+USE testdb;
+
+CREATE TABLE t1(id INT, name STRING, value DOUBLE) USING PARQUET;
+CREATE TABLE t2(id INT, category STRING, amount BIGINT) USING CSV;
+CREATE TABLE t3(a INT, b STRING, c DOUBLE) USING ORC;
+
+-- Basic metadata column selection
+SELECT _metadata.file_name FROM t1;
+SELECT _metadata.file_path FROM t1;
+SELECT _metadata.file_size FROM t1;
+SELECT _metadata.file_modification_time FROM t1;
+SELECT _metadata.row_index FROM t1;
+SELECT _metadata.file_block_start FROM t1;
+SELECT _metadata.file_block_length FROM t1;
+
+-- Qualified table references with metadata
+SELECT t1._metadata.file_name FROM t1;
+SELECT testdb.t1._metadata.file_name FROM testdb.t1;
+
+-- Mixed qualified and unqualified metadata references
+SELECT _metadata.file_name, t1._metadata.file_path FROM t1;
+SELECT t1._metadata.file_name, testdb.t1._metadata.file_path FROM testdb.t1;
+
+-- Metadata in WHERE clause
+SELECT * FROM t1 WHERE _metadata.file_size > 1000;
+SELECT id, name FROM t1 WHERE _metadata.row_index < 100;
+SELECT * FROM t1 WHERE _metadata.file_name = 'data.parquet';
+SELECT * FROM t1 WHERE _metadata.file_modification_time > timestamp'2024-01-01 00:00:00';
+
+-- Qualified metadata in WHERE clause
+SELECT * FROM t1 WHERE t1._metadata.file_size > 1000;
+SELECT * FROM testdb.t1 WHERE testdb.t1._metadata.row_index < 100;
+
+-- Metadata in ORDER BY clause
+SELECT id, name, _metadata.file_name FROM t1 ORDER BY _metadata.file_size;
+SELECT * FROM t1 ORDER BY _metadata.row_index DESC;
+SELECT id, _metadata.file_name FROM t1 ORDER BY _metadata.file_modification_time, _metadata.row_index;
+
+-- Qualified metadata in GROUP BY
+SELECT t1._metadata.file_name, COUNT(*) FROM t1 GROUP BY t1._metadata.file_name;
+SELECT testdb.t1._metadata.file_path, SUM(value) FROM testdb.t1 GROUP BY testdb.t1._metadata.file_path;
+
+-- Metadata with aggregations
+SELECT _metadata.file_name, MIN(_metadata.row_index), MAX(_metadata.row_index) FROM t1
+GROUP BY _metadata.file_name;
+SELECT _metadata.file_path, COUNT(DISTINCT _metadata.file_name) FROM t1
+GROUP BY _metadata.file_path;
+
+-- Metadata in JOIN conditions
+SELECT t1.id, t2.category, t1._metadata.file_name
+FROM t1 JOIN t2 ON t1.id = t2.id;
+
+SELECT t1.id, t1._metadata.file_name, t2._metadata.file_name
+FROM t1 JOIN t2 ON t1.id = t2.id;
+
+SELECT t1.id, t1._metadata.row_index, t2._metadata.row_index
+FROM t1 JOIN t2 ON t1.id = t2.id
+WHERE t1._metadata.file_size > 1000;
+
+-- Metadata in LEFT/RIGHT JOIN
+SELECT t1.name, t1._metadata.file_name, t2._metadata.file_name
+FROM t1 LEFT JOIN t2 ON t1.id = t2.id;
+
+SELECT t1._metadata.file_path, t2._metadata.file_path, t1.id
+FROM t1 RIGHT JOIN t2 ON t1.id = t2.id;
+
+-- Metadata in FULL OUTER JOIN
+SELECT t1._metadata.file_name, t2._metadata.file_name, t1.id, t2.id
+FROM t1 FULL OUTER JOIN t2 ON t1.id = t2.id;
+
+-- Metadata in self-join
+SELECT a.id, a._metadata.file_name, b._metadata.file_name
+FROM t1 a JOIN t1 b ON a.id = b.id
+WHERE a._metadata.row_index < b._metadata.row_index;
+
+-- Metadata with table aliases
+SELECT t._metadata.file_name, t.id FROM t1 t;
+SELECT x._metadata.file_name, x._metadata.row_index FROM t1 x
+WHERE x._metadata.file_size > 500;
+
+-- Metadata in subqueries
+SELECT * FROM (SELECT id, _metadata.file_name FROM t1) sub;
+SELECT id FROM t1 WHERE _metadata.file_name IN
+ (SELECT _metadata.file_name FROM t2 WHERE amount > 100);
+
+-- Metadata in UNION
+SELECT _metadata.file_name, id FROM t1
+UNION
+SELECT _metadata.file_name, id FROM t2;
+
+-- === _metadata.* struct expansion ===
+
+-- expand all metadata fields for each format
+SELECT _metadata.* FROM t1;
+SELECT _metadata.* FROM t2;
+SELECT _metadata.* FROM t3;
+
+-- qualified _metadata.* expansion
+SELECT t1._metadata.* FROM t1;
+SELECT t2._metadata.* FROM t2;
+SELECT t3._metadata.* FROM t3;
+
+-- _metadata.* alongside regular columns (parquet)
+SELECT id, name, _metadata.* FROM t1 ORDER BY id;
+
+-- _metadata.* alongside regular columns (csv)
+SELECT id, category, _metadata.* FROM t2 ORDER BY id;
+
+-- _metadata.* in subquery
+SELECT sub.file_name, sub.file_size FROM (SELECT _metadata.* FROM t1) sub ORDER BY sub.file_name;
+
+-- _metadata.* with WHERE on expanded fields
+SELECT id, _metadata.* FROM t1 WHERE _metadata.file_size >= 0 ORDER BY id;
+
+-- _metadata.* in Aggregate
+SELECT _metadata.*, COUNT(*) AS cnt FROM t1 GROUP BY ALL ORDER BY _metadata.file_name;
+SELECT _metadata.*, COUNT(*) AS cnt FROM t2 GROUP BY ALL ORDER BY _metadata.file_name;
+SELECT _metadata.*, COUNT(*) AS cnt FROM t3 GROUP BY ALL ORDER BY _metadata.file_name;
+
+-- Cleanup
+DROP TABLE t1;
+DROP TABLE t2;
+DROP TABLE t3;
+DROP DATABASE testdb;
diff --git a/sql/core/src/test/resources/sql-tests/results/metadata_column_resolution.sql.out b/sql/core/src/test/resources/sql-tests/results/metadata_column_resolution.sql.out
new file mode 100644
index 000000000000..d9a1176a8c8e
--- /dev/null
+++ b/sql/core/src/test/resources/sql-tests/results/metadata_column_resolution.sql.out
@@ -0,0 +1,493 @@
+-- Automatically generated by SQLQueryTestSuite
+-- !query
+CREATE DATABASE IF NOT EXISTS testdb
+-- !query schema
+struct<>
+-- !query output
+
+
+
+-- !query
+USE testdb
+-- !query schema
+struct<>
+-- !query output
+
+
+
+-- !query
+CREATE TABLE t1(id INT, name STRING, value DOUBLE) USING PARQUET
+-- !query schema
+struct<>
+-- !query output
+
+
+
+-- !query
+CREATE TABLE t2(id INT, category STRING, amount BIGINT) USING CSV
+-- !query schema
+struct<>
+-- !query output
+
+
+
+-- !query
+CREATE TABLE t3(a INT, b STRING, c DOUBLE) USING ORC
+-- !query schema
+struct<>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.file_name FROM t1
+-- !query schema
+struct<file_name:string>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.file_path FROM t1
+-- !query schema
+struct<file_path:string>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.file_size FROM t1
+-- !query schema
+struct<file_size:bigint>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.file_modification_time FROM t1
+-- !query schema
+struct<file_modification_time:timestamp>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.row_index FROM t1
+-- !query schema
+struct<row_index:bigint>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.file_block_start FROM t1
+-- !query schema
+struct<file_block_start:bigint>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.file_block_length FROM t1
+-- !query schema
+struct<file_block_length:bigint>
+-- !query output
+
+
+
+-- !query
+SELECT t1._metadata.file_name FROM t1
+-- !query schema
+struct<file_name:string>
+-- !query output
+
+
+
+-- !query
+SELECT testdb.t1._metadata.file_name FROM testdb.t1
+-- !query schema
+struct<file_name:string>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.file_name, t1._metadata.file_path FROM t1
+-- !query schema
+struct<file_name:string,file_path:string>
+-- !query output
+
+
+
+-- !query
+SELECT t1._metadata.file_name, testdb.t1._metadata.file_path FROM testdb.t1
+-- !query schema
+struct<file_name:string,file_path:string>
+-- !query output
+
+
+
+-- !query
+SELECT * FROM t1 WHERE _metadata.file_size > 1000
+-- !query schema
+struct<id:int,name:string,value:double>
+-- !query output
+
+
+
+-- !query
+SELECT id, name FROM t1 WHERE _metadata.row_index < 100
+-- !query schema
+struct<id:int,name:string>
+-- !query output
+
+
+
+-- !query
+SELECT * FROM t1 WHERE _metadata.file_name = 'data.parquet'
+-- !query schema
+struct<id:int,name:string,value:double>
+-- !query output
+
+
+
+-- !query
+SELECT * FROM t1 WHERE _metadata.file_modification_time > timestamp'2024-01-01 00:00:00'
+-- !query schema
+struct<id:int,name:string,value:double>
+-- !query output
+
+
+
+-- !query
+SELECT * FROM t1 WHERE t1._metadata.file_size > 1000
+-- !query schema
+struct<id:int,name:string,value:double>
+-- !query output
+
+
+
+-- !query
+SELECT * FROM testdb.t1 WHERE testdb.t1._metadata.row_index < 100
+-- !query schema
+struct<id:int,name:string,value:double>
+-- !query output
+
+
+
+-- !query
+SELECT id, name, _metadata.file_name FROM t1 ORDER BY _metadata.file_size
+-- !query schema
+struct<id:int,name:string,file_name:string>
+-- !query output
+
+
+
+-- !query
+SELECT * FROM t1 ORDER BY _metadata.row_index DESC
+-- !query schema
+struct<id:int,name:string,value:double>
+-- !query output
+
+
+
+-- !query
+SELECT id, _metadata.file_name FROM t1 ORDER BY _metadata.file_modification_time, _metadata.row_index
+-- !query schema
+struct<id:int,file_name:string>
+-- !query output
+
+
+
+-- !query
+SELECT t1._metadata.file_name, COUNT(*) FROM t1 GROUP BY t1._metadata.file_name
+-- !query schema
+struct<file_name:string,count(1):bigint>
+-- !query output
+
+
+
+-- !query
+SELECT testdb.t1._metadata.file_path, SUM(value) FROM testdb.t1 GROUP BY testdb.t1._metadata.file_path
+-- !query schema
+struct<file_path:string,sum(value):double>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.file_name, MIN(_metadata.row_index), MAX(_metadata.row_index) FROM t1
+GROUP BY _metadata.file_name
+-- !query schema
+struct<file_name:string,min(_metadata.row_index):bigint,max(_metadata.row_index):bigint>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.file_path, COUNT(DISTINCT _metadata.file_name) FROM t1
+GROUP BY _metadata.file_path
+-- !query schema
+struct<file_path:string,count(DISTINCT _metadata.file_name):bigint>
+-- !query output
+
+
+
+-- !query
+SELECT t1.id, t2.category, t1._metadata.file_name
+FROM t1 JOIN t2 ON t1.id = t2.id
+-- !query schema
+struct<id:int,category:string,file_name:string>
+-- !query output
+
+
+
+-- !query
+SELECT t1.id, t1._metadata.file_name, t2._metadata.file_name
+FROM t1 JOIN t2 ON t1.id = t2.id
+-- !query schema
+struct<id:int,file_name:string,file_name:string>
+-- !query output
+
+
+
+-- !query
+SELECT t1.id, t1._metadata.row_index, t2._metadata.row_index
+FROM t1 JOIN t2 ON t1.id = t2.id
+WHERE t1._metadata.file_size > 1000
+-- !query schema
+struct<>
+-- !query output
+org.apache.spark.sql.AnalysisException
+{
+ "errorClass" : "FIELD_NOT_FOUND",
+ "sqlState" : "42704",
+ "messageParameters" : {
+ "fieldName" : "`row_index`",
+ "fields" : "`file_path`, `file_name`, `file_size`, `file_block_start`, `file_block_length`, `file_modification_time`"
+ },
+ "queryContext" : [ {
+ "objectType" : "",
+ "objectName" : "",
+ "startIndex" : 39,
+ "stopIndex" : 60,
+ "fragment" : "t2._metadata.row_index"
+ } ]
+}
+
+
+-- !query
+SELECT t1.name, t1._metadata.file_name, t2._metadata.file_name
+FROM t1 LEFT JOIN t2 ON t1.id = t2.id
+-- !query schema
+struct<name:string,file_name:string,file_name:string>
+-- !query output
+
+
+
+-- !query
+SELECT t1._metadata.file_path, t2._metadata.file_path, t1.id
+FROM t1 RIGHT JOIN t2 ON t1.id = t2.id
+-- !query schema
+struct<file_path:string,file_path:string,id:int>
+-- !query output
+
+
+
+-- !query
+SELECT t1._metadata.file_name, t2._metadata.file_name, t1.id, t2.id
+FROM t1 FULL OUTER JOIN t2 ON t1.id = t2.id
+-- !query schema
+struct<file_name:string,file_name:string,id:int,id:int>
+-- !query output
+
+
+
+-- !query
+SELECT a.id, a._metadata.file_name, b._metadata.file_name
+FROM t1 a JOIN t1 b ON a.id = b.id
+WHERE a._metadata.row_index < b._metadata.row_index
+-- !query schema
+struct<id:int,file_name:string,file_name:string>
+-- !query output
+
+
+
+-- !query
+SELECT t._metadata.file_name, t.id FROM t1 t
+-- !query schema
+struct<file_name:string,id:int>
+-- !query output
+
+
+
+-- !query
+SELECT x._metadata.file_name, x._metadata.row_index FROM t1 x
+WHERE x._metadata.file_size > 500
+-- !query schema
+struct<file_name:string,row_index:bigint>
+-- !query output
+
+
+
+-- !query
+SELECT * FROM (SELECT id, _metadata.file_name FROM t1) sub
+-- !query schema
+struct<id:int,file_name:string>
+-- !query output
+
+
+
+-- !query
+SELECT id FROM t1 WHERE _metadata.file_name IN
+ (SELECT _metadata.file_name FROM t2 WHERE amount > 100)
+-- !query schema
+struct<id:int>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.file_name, id FROM t1
+UNION
+SELECT _metadata.file_name, id FROM t2
+-- !query schema
+struct<file_name:string,id:int>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.* FROM t1
+-- !query schema
+struct<file_path:string,file_name:string,file_size:bigint,file_block_start:bigint,file_block_length:bigint,file_modification_time:timestamp,row_index:bigint>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.* FROM t2
+-- !query schema
+struct<file_path:string,file_name:string,file_size:bigint,file_block_start:bigint,file_block_length:bigint,file_modification_time:timestamp>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.* FROM t3
+-- !query schema
+struct<file_path:string,file_name:string,file_size:bigint,file_block_start:bigint,file_block_length:bigint,file_modification_time:timestamp>
+-- !query output
+
+
+
+-- !query
+SELECT t1._metadata.* FROM t1
+-- !query schema
+struct<file_path:string,file_name:string,file_size:bigint,file_block_start:bigint,file_block_length:bigint,file_modification_time:timestamp,row_index:bigint>
+-- !query output
+
+
+
+-- !query
+SELECT t2._metadata.* FROM t2
+-- !query schema
+struct<file_path:string,file_name:string,file_size:bigint,file_block_start:bigint,file_block_length:bigint,file_modification_time:timestamp>
+-- !query output
+
+
+
+-- !query
+SELECT t3._metadata.* FROM t3
+-- !query schema
+struct<file_path:string,file_name:string,file_size:bigint,file_block_start:bigint,file_block_length:bigint,file_modification_time:timestamp>
+-- !query output
+
+
+
+-- !query
+SELECT id, name, _metadata.* FROM t1 ORDER BY id
+-- !query schema
+struct<id:int,name:string,file_path:string,file_name:string,file_size:bigint,file_block_start:bigint,file_block_length:bigint,file_modification_time:timestamp,row_index:bigint>
+-- !query output
+
+
+
+-- !query
+SELECT id, category, _metadata.* FROM t2 ORDER BY id
+-- !query schema
+struct<id:int,category:string,file_path:string,file_name:string,file_size:bigint,file_block_start:bigint,file_block_length:bigint,file_modification_time:timestamp>
+-- !query output
+
+
+
+-- !query
+SELECT sub.file_name, sub.file_size FROM (SELECT _metadata.* FROM t1) sub ORDER BY sub.file_name
+-- !query schema
+struct<file_name:string,file_size:bigint>
+-- !query output
+
+
+
+-- !query
+SELECT id, _metadata.* FROM t1 WHERE _metadata.file_size >= 0 ORDER BY id
+-- !query schema
+struct<id:int,file_path:string,file_name:string,file_size:bigint,file_block_start:bigint,file_block_length:bigint,file_modification_time:timestamp,row_index:bigint>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.*, COUNT(*) AS cnt FROM t1 GROUP BY ALL ORDER BY _metadata.file_name
+-- !query schema
+struct<file_path:string,file_name:string,file_size:bigint,file_block_start:bigint,file_block_length:bigint,file_modification_time:timestamp,row_index:bigint,cnt:bigint>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.*, COUNT(*) AS cnt FROM t2 GROUP BY ALL ORDER BY _metadata.file_name
+-- !query schema
+struct<file_path:string,file_name:string,file_size:bigint,file_block_start:bigint,file_block_length:bigint,file_modification_time:timestamp,cnt:bigint>
+-- !query output
+
+
+
+-- !query
+SELECT _metadata.*, COUNT(*) AS cnt FROM t3 GROUP BY ALL ORDER BY _metadata.file_name
+-- !query schema
+struct<file_path:string,file_name:string,file_size:bigint,file_block_start:bigint,file_block_length:bigint,file_modification_time:timestamp,cnt:bigint>
+-- !query output
+
+
+
+-- !query
+DROP TABLE t1
+-- !query schema
+struct<>
+-- !query output
+
+
+
+-- !query
+DROP TABLE t2
+-- !query schema
+struct<>
+-- !query output
+
+
+
+-- !query
+DROP TABLE t3
+-- !query schema
+struct<>
+-- !query output
+
+
+
+-- !query
+DROP DATABASE testdb
+-- !query schema
+struct<>
+-- !query output
+
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@spark.apache.org
For additional commands, e-mail: commits-help@spark.apache.org