IDENTIFIER clause

Description

Converts a constant STRING expression into a SQL object name. The purpose of this clause is to allow for templating of identifiers in SQL statements without opening up the risk of SQL injection attacks.

The clause comes in two forms:

When passed a string-literal, which may include a coalesced string of literals and parameter markers, it can be used anywhere an identifier or qualified identifier can be used. The usage of this form is encouraged.
When passed a more complex constant string expression, which may also include variables, it can be used in limited cases to reference dynamic SQL object references such as table names, column names, and function names. The usage of this form is discouraged unless necessary. For example you cannot use this form to parameterize a table or column alias, or a column name in a CREATE TABLE statement.

Syntax

IDENTIFIER ( strLiteral )

IDENTIFIER ( strExpr )

Parameters

strLiteral

A constant STRING literal which may include string coalescing of strings and string parameter markers.

The string must be a valid (qualified) identifier.

For example:
```
IDENTIFIER('default.' :tablename)
IDENTIFIER(:catalog '.' :namespace '.' :tablename '.' :columnname)
IDENTIFIER(:rootdirectory '/data/' :tablename)
```

strExpr: A constant STRING expression. Typically, the expression includes a parameter marker or session variable.

The string must be a valid (qualified) identifier.

For example:

IDENTIFIER(session.schema || '.' || session.tablename)
IDENTIFIER('`' || session.schema || '`.`' session.tablename || '`')
IDENTIFIER(CONCAT(:catalog, '.', :namespace, '.', :tablename, '.', :columnname))

Returns

A (qualified) identifier.

Examples

Scala examples

These examples use named parameter markers to templatize queries.

// Creation of a table using parameter marker.
spark.sql("CREATE TABLE IDENTIFIER(:mytab)(c1 INT)", args = Map("mytab" -> "tab1")).show()

spark.sql("DESCRIBE IDENTIFIER(:mytab)", args = Map("mytab" -> "tab1")).show()
+--------+---------+-------+
|col_name|data_type|comment|
+--------+---------+-------+
|      c1|      int|   NULL|
+--------+---------+-------+

// Altering a table with a fixed schema and a parameterized table name. 
spark.sql("ALTER TABLE IDENTIFIER('default.' || :mytab) ADD COLUMN c2 INT", args = Map("mytab" -> "tab1")).show()

spark.sql("DESCRIBE IDENTIFIER(:mytab)", args = Map("mytab" -> "default.tab1")).show()
+--------+---------+-------+
|col_name|data_type|comment|
+--------+---------+-------+
|      c1|      int|   NULL|
|      c2|      int|   NULL|
+--------+---------+-------+

// A parameterized reference to a table in a query. This table name is qualified and uses back-ticks.
spark.sql("SELECT * FROM IDENTIFIER(:mytab)", args = Map("mytab" -> "`default`.`tab1`")).show()
+---+---+
| c1| c2|
+---+---+
+---+---+


// You cannot qualify the IDENTIFIER clause or use it as a qualifier itself.
spark.sql("SELECT * FROM myschema.IDENTIFIER(:mytab)", args = Map("mytab" -> "`tab1`")).show()
[INVALID_SQL_SYNTAX.INVALID_TABLE_VALUED_FUNC_NAME] `myschema`.`IDENTIFIER`.

spark.sql("SELECT * FROM IDENTIFIER(:myschema).mytab", args = Map("mychema" -> "`default`")).show()
[PARSE_SYNTAX_ERROR]

// Dropping a table with separate schema and table parameters.
spark.sql("DROP TABLE IDENTIFIER(:myschema '.' :mytab)", args = Map("myschema" -> "default", "mytab" -> "tab1")).show()

// A parameterized column reference
spark.sql("SELECT IDENTIFIER(:col) FROM VALUES(1) AS T(c1)", args = Map("col" -> "t.c1")).show()
+---+
| c1|
+---+
|  1|
+---+

// Passing in a function name as a parameter
spark.sql("SELECT IDENTIFIER(:func)(-1)", args = Map("func" -> "abs")).show();
+-------+
|abs(-1)|
+-------+
|      1|
+-------+

SQL examples

These examples use SQL variables to templatize queries.

DECLARE mytab = 'tab1';

-- Creation of a table using variable.
CREATE TABLE IDENTIFIER(mytab)(c1 INT);

EXECUTE IMMEDIATE 'DESCRIBE IDENTIFIER(:mytab)' USING mytab;
+--------+---------+-------+
|col_name|data_type|comment|
+--------+---------+-------+
|      c1|      int|   NULL|
+--------+---------+-------+

-- Altering a table with a fixed schema and a parameterized table name. 
EXECUTE IMMEDIATE 'ALTER TABLE IDENTIFIER('default.' || :mytab) ADD COLUMN :col INT' USING mytab, 'c2' AS col;

SET VAR mytab = '`default`.`tab1`';
EXECUTE IMMEDIATE 'DESCRIBE IDENTIFIER(:mytab)' USING mytab;
+--------+---------+-------+
|col_name|data_type|comment|
+--------+---------+-------+
|      c1|      int|   NULL|
|      c2|      int|   NULL|
+--------+---------+-------+

-- A parameterized reference to a table in a query. This table name is qualified and uses back-ticks.
EXECUTE IMMEDIATE 'SELECT * FROM IDENTIFIER(mytab)' USING mytab;
+---+---+
| c1| c2|
+---+---+

-- Dropping a table with separate schema and table parameters.
DECLARE myschema = 'default';
SET VAR mytab = 'tab1';
EXECUTE IMMEDIATE 'DROP TABLE IDENTIFIER(:myschema '.' :mytab)' USING myschema, mytab;

-- A parameterized column reference
DECLARE col = 'c1';
EXECUTE IMEMDIATE 'SELECT IDENTIFIER(:col) FROM VALUES(1) AS T(IDENTIFIER(:col))' USING col;
+---+
| c1|
+---+
|  1|
+---+

-- Passing in a function name as a parameter
EXECUTE IMMEDIATE 'SELECT IDENTIFIER(:func)(-1)' USING 'abs' AS func;
+-------+
|abs(-1)|
+-------+
|      1|
+-------+

Spark SQL Guide