spark-instrumented-optimizer/sql/catalyst
Max Gekk 2afd1fb492 [SPARK-33904][SQL] Recognize spark_catalog in saveAsTable() and insertInto()
### What changes were proposed in this pull request?
In the `saveAsTable()` and `insertInto()` methods of `DataFrameWriter`, recognize `spark_catalog` as the default session catalog in table names.

### Why are the changes needed?
1. To simplify writing of unified v1 and v2 tests
2. To improve Spark SQL user experience. `insertInto()` should have feature parity with the `INSERT INTO` sql command. Currently, `insertInto()` fails on a table from a namespace in `spark_catalog`:
```scala
scala> sql("CREATE NAMESPACE spark_catalog.ns")
scala> Seq(0).toDF().write.saveAsTable("spark_catalog.ns.tbl")
org.apache.spark.sql.AnalysisException: Couldn't find a catalog to handle the identifier spark_catalog.ns.tbl.
  at org.apache.spark.sql.DataFrameWriter.saveAsTable(DataFrameWriter.scala:629)
  ... 47 elided
scala> Seq(0).toDF().write.insertInto("spark_catalog.ns.tbl")
org.apache.spark.sql.AnalysisException: Couldn't find a catalog to handle the identifier spark_catalog.ns.tbl.
  at org.apache.spark.sql.DataFrameWriter.insertInto(DataFrameWriter.scala:498)
  ... 47 elided
```
but `INSERT INTO` succeed:
```sql
spark-sql> create table spark_catalog.ns.tbl (c int);
spark-sql> insert into spark_catalog.ns.tbl select 0;
spark-sql> select * from spark_catalog.ns.tbl;
0
```

### Does this PR introduce _any_ user-facing change?
Yes. After the changes for the example above:
```scala
scala> Seq(0).toDF().write.saveAsTable("spark_catalog.ns.tbl")
scala> Seq(1).toDF().write.insertInto("spark_catalog.ns.tbl")
scala> spark.table("spark_catalog.ns.tbl").show(false)
+-----+
|value|
+-----+
|0    |
|1    |
+-----+
```

### How was this patch tested?
By running the affected test suites:
```
$ build/sbt -Phive-2.3 -Phive-thriftserver "test:testOnly *.ShowPartitionsSuite"
$ build/sbt -Phive-2.3 -Phive-thriftserver "test:testOnly *.FileFormatWriterSuite"
```

Closes #30919 from MaxGekk/insert-into-spark_catalog.

Authored-by: Max Gekk <max.gekk@gmail.com>
Signed-off-by: Wenchen Fan <wenchen@databricks.com>
2020-12-30 07:56:34 +00:00
..
benchmarks [SPARK-30413][SQL] Avoid WrappedArray roundtrip in GenericArrayData constructor, plus related optimization in ParquetMapConverter 2020-01-19 19:12:19 -08:00
src [SPARK-33904][SQL] Recognize spark_catalog in saveAsTable() and insertInto() 2020-12-30 07:56:34 +00:00
pom.xml [SPARK-33662][BUILD] Setting version to 3.2.0-SNAPSHOT 2020-12-04 14:10:42 -08:00