SELECT Statements

A SELECT statement can consist of the following basic clauses. You can access this statement by using the Odbc.Query function in the M formula language.

SELECT
INTO
FROM
JOIN
WHERE
GROUP BY
HAVING
UNION
ORDER BY
LIMIT

SELECT Syntax

The following syntax diagram outlines the syntax supported by the SQL engine of the connector:

SELECT {
  [ TOP <numeric_literal> | DISTINCT ]
  { 
    * 
    | { 
        <expression> [ [ AS ] <column_reference> ] 
        | { <table_name> | <correlation_name> } .* 
      } [ , ... ] 
  }
  { 
    FROM <table_reference> [ [ AS ] <identifier> ] 
  } [ , ... ]
  [ [  
      INNER | { { LEFT | RIGHT | FULL } [ OUTER ] } 
    ] JOIN <table_reference> [ ON <search_condition> ] [ [ AS ] <identifier> ] 
  ] [ ... ] 
  [ WHERE <search_condition> ]
  [ GROUP BY <column_reference> [ , ... ]
  [ HAVING <search_condition> ]
  [ UNION [ ALL ] <select_statement> ]
  [ 
    ORDER BY 
    <column_reference> [ ASC | DESC ] [ NULLS FIRST | NULLS LAST ]
  ]
  [ 
    LIMIT <expression>
    [ 
      { OFFSET | , }
      <expression> 
    ]
  ] 
} | SCOPE_IDENTITY() 

<expression> ::=
  | <column_reference>
  | @ <parameter> 
  | ?
  | COUNT( * | { [ DISTINCT ] <expression> } )
  | { AVG | MAX | MIN | SUM | COUNT } ( <expression> ) 
  | NULLIF ( <expression> , <expression> ) 
  | COALESCE ( <expression> , ... ) 
  | CASE <expression>
      WHEN { <expression> | <search_condition> } THEN { <expression> | NULL } [ ... ]
    [ ELSE { <expression> | NULL } ]
    END 
  | {RANK() | DENSE_RANK()} OVER ([PARTITION BY <column_reference>] {ORDER BY <column_reference>})
  | <literal>
  | <sql_function> 

<search_condition> ::= 
  {
    <expression> { = | > | < | >= | <= | <> | != | LIKE | NOT LIKE | IN | NOT IN | IS NULL | IS NOT NULL | AND | OR | CONTAINS | BETWEEN } [ <expression> ]
  } [ { AND | OR } ... ]

Examples

Return all columns:
```
SELECT * FROM Customers
```

Rename a column:

SELECT [CompanyName] AS MY_CompanyName FROM Customers

Cast a column's data as a different data type:

SELECT CAST(Balance AS VARCHAR) AS Str_Balance FROM Customers

Search data:

SELECT * FROM Customers WHERE Country = 'US'

Return the number of items matching the query criteria:
```
SELECT COUNT(*) AS MyCount FROM Customers 
```
Return the number of unique items matching the query criteria:
```
SELECT COUNT(DISTINCT CompanyName) FROM Customers 
```
Return the unique items matching the query criteria:
```
SELECT DISTINCT CompanyName FROM Customers 
```

Sort a result set in ascending order:

SELECT City, CompanyName FROM Customers  ORDER BY CompanyName ASC

Restrict a result set to the specified number of rows:
```
SELECT City, CompanyName FROM Customers LIMIT 10 
```
Parameterize a query to pass in inputs at execution time. This enables you to create prepared statements and mitigate SQL injection attacks.
```
SELECT * FROM Customers WHERE Country = @param
```

Pseudo Columns

Some input-only fields are available in SELECT statements. These fields, called pseudo columns, do not appear as regular columns in the results, yet may be specified as part of the WHERE clause. You can use pseudo columns to access additional features from Spark SQL.

    SELECT * FROM Customers WHERE MyPseudocolumn = 'MyValue'

Aggregate Functions

For SELECT examples using aggregate functions, see Aggregate Functions.

JOIN Queries

See JOIN Queries for SELECT query examples using JOINs.

Date Literal Functions

Date Literal Functions contains SELECT examples with date literal functions.

Projection Functions

See Projection Functions for SELECT examples with projection functions.

Predicate Functions

For SELECT examples using predicate functions, see Predicate Functions.

Power BI Connector for Spark SQL