SQL language reference

This is a SQL command reference for Databricks SQL and Databricks Runtime.

For information about how to understand and use the syntax notation and symbols in this reference, see How to use the SQL reference.

For information about using SQL with Lakeflow Spark Declarative Pipelines, see Pipeline SQL language reference.

General reference

This general reference describes data types, functions, identifiers, literals, and semantics.

Language fundamentals

Reference	Description
Adding comments to SQL statements	Adding comments to your SQL code
Reserved words and schemas	Keywords reserved by the SQL parser
Identifiers	Rules for naming database objects
Names	Object naming conventions
IDENTIFIER clause	Dynamic identifier resolution
SQL expression	SQL expression syntax and evaluation
NULL semantics	How NULL values are handled
Parameter markers	Parameterized query placeholders
Variables	Session and temporary variables
Name resolution	How object names are resolved
JSON path expression	Navigating JSON data structures
Collation	String comparison and sorting rules
Partitions	Table partitioning concepts

Data types

Reference	Description
Data types	Overview of all supported data types
SQL data type rules	Type coercion and conversion rules
Datetime patterns	Format patterns for dates and times

Configuration parameters

Reference	Description
Configuration parameters	Overview of SQL configuration
ANSI_MODE	ANSI SQL compliance mode
LEGACY_TIME_PARSER_POLICY	Time parsing behavior
MAX_FILE_PARTITION_BYTES	File partition size limit
READ_ONLY_EXTERNAL_METASTORE	External metastore access mode
STATEMENT_TIMEOUT	Query timeout setting
TIMEZONE	Session timezone
USE_CACHED_RESULT	Query result caching

Functions

Reference	Description
Functions	Overview of SQL functions
Built-in functions	All built-in functions
Alphabetical list of built-in functions	Complete A-Z function reference
Window functions	Analytic and ranking functions
Lambda functions	Anonymous function expressions
H3 geospatial functions	Hexagonal hierarchical spatial index
ST geospatial functions	Standard geospatial functions
User-defined aggregate functions (UDAFs)	Custom aggregation functions
External user-defined scalar functions (UDFs)	Custom scalar functions
Integration with Hive UDFs, UDAFs, and UDTFs	Hive function compatibility
Function invocation	How to call functions

Unity Catalog concepts

Reference	Description
Principal	Users, groups, and service principals
Privileges and securable objects in Unity Catalog	Access control in Unity Catalog
Privileges and securable objects in the Hive metastore	Legacy metastore access control
External locations	Cloud storage locations
External tables	Tables backed by external data
Credentials	Storage access credentials
Volumes	Managed and external file storage

Special topics

Reference	Description
ANSI compliance in Databricks Runtime	ANSI SQL standard compatibility
Apache Hive compatibility	Hive SQL compatibility
SQL scripting	Procedural SQL logic
Delta Sharing	Cross-organization data sharing
Federated queries (Lakehouse Federation)	Querying external databases
Information schema	Metadata system tables

DDL statements

Data definition statements create or modify the structure of database objects.

ALTER statements

Statement	Description
ALTER CATALOG	Modify catalog properties
ALTER CONNECTION	Modify connection settings
ALTER CREDENTIAL	Modify credential properties
ALTER DATABASE	Modify database properties
ALTER EXTERNAL LOCATION	Modify external location
ALTER MATERIALIZED VIEW	Modify materialized view
ALTER PROVIDER	Modify data provider
ALTER RECIPIENT	Modify share recipient
ALTER SCHEMA	Modify schema properties
ALTER SHARE	Modify share configuration
ALTER STREAMING TABLE	Modify streaming table
ALTER TABLE	Modify table structure
ALTER VIEW	Modify view definition
ALTER VOLUME	Modify volume properties

CREATE statements

Statement	Description
CREATE BLOOM FILTER INDEX	Create bloom filter index
CREATE CATALOG	Create a new catalog
CREATE CONNECTION	Create external connection
CREATE DATABASE	Create a new database
CREATE FUNCTION (SQL and Python)	Create SQL function
CREATE FUNCTION (External)	Create external function
CREATE EXTERNAL LOCATION	Create external location
CREATE MATERIALIZED VIEW	Create materialized view
CREATE PROCEDURE	Create stored procedure
CREATE RECIPIENT	Create share recipient
CREATE SCHEMA	Create a new schema
CREATE SERVER	Create foreign server
CREATE SHARE	Create data share
CREATE STREAMING TABLE	Create streaming table
CREATE TABLE	Create a new table
CREATE VIEW	Create a view
CREATE VOLUME	Create a volume

DROP statements

• Analyze statements
• Other DDL statements

Analyze statements

• ANALYZE TABLE … COMPUTE STATISTICS
• ANALYZE TABLE … COMPUTE STORAGE METRICS

Other DDL statements

Statement	Description
COMMENT ON	Add comments to objects
DECLARE VARIABLE	Declare session variable
REPAIR TABLE	Recover table partitions
REFRESH FOREIGN (CATALOG, SCHEMA, and TABLE)	Refresh foreign metadata
REFRESH (MATERIALIZED VIEW or STREAMING TABLE)	Refresh materialized data
SET TAG	Add tags to objects
SYNC	Synchronize metadata
TRUNCATE TABLE	Remove all table data
UNDROP	Restore dropped table
UNSET TAG	Remove tags from objects

DML statements

Data manipulation statements add, change, or delete data in Delta Lake tables.

Statement	Description
COPY INTO	Load data from files into a table
DELETE FROM	Delete rows from a table
INSERT	Insert rows into a table
INSERT OVERWRITE DIRECTORY	Write query results to directory
INSERT OVERWRITE DIRECTORY with Hive format	Write results in Hive format
LOAD DATA	Load data from files
MERGE INTO	Upsert data into a table
UPDATE	Update rows in a table

Data retrieval statements

Retrieve rows from tables according to specified clauses. Databricks supports both standard SELECT syntax and SQL Pipeline Syntax.

Query statements

Statement	Description
Query	Full query syntax reference
SELECT (subselect)	Select data from tables
VALUES clause	Construct inline table
SQL Pipeline Syntax	Pipe-based query composition
EXPLAIN	Show query execution plan

Query clauses

Clause	Description
SELECT clause	Column selection
* (star) clause	Select all columns
table reference	Table and subquery sources
JOIN	Combine rows from tables
WHERE clause	Filter rows
GROUP BY clause	Group rows for aggregation
HAVING clause	Filter grouped rows
QUALIFY clause	Filter window function results
ORDER BY clause	Sort result rows
SORT BY clause	Sort within partitions
CLUSTER BY clause (SELECT)	Distribute and sort data
DISTRIBUTE BY clause	Distribute data across partitions
LIMIT clause	Limit result rows
OFFSET clause	Skip result rows
PIVOT clause	Rotate rows to columns
UNPIVOT clause	Rotate columns to rows
LATERAL VIEW clause	Apply table-generating function
TABLESAMPLE clause	Sample table rows
Table-valued function (TVF) invocation	Functions returning tables
Common table expression (CTE)	Named subqueries
Set operators	UNION, INTERSECT, EXCEPT
WINDOW clause	Named window definitions
Hints	Query optimizer hints

Delta Lake statements

Manage tables stored in Delta Lake format. For details, see What is Delta Lake in Databricks?.

Statement	Description
CACHE SELECT	Cache query results
CONVERT TO DELTA	Convert Parquet to Delta
DESCRIBE HISTORY	Show table history
FSCK REPAIR TABLE	Repair table metadata
GENERATE	Generate table manifest
OPTIMIZE	Compact table files
REORG TABLE	Reorganize table data
RESTORE	Restore table to previous version
VACUUM	Remove old table files

SQL scripting statements

Execute procedural logic in SQL. See SQL scripting for an overview.

Statement	Description
BEGIN END compound statement	Group statements into a block
CASE statement	Conditional branching
FOR statement	Loop over a range or cursor
GET DIAGNOSTICS statement	Retrieve execution information
IF THEN ELSE statement	Conditional execution
ITERATE statement	Continue to next loop iteration
LEAVE statement	Exit a loop or block
LOOP statement	Unconditional loop
REPEAT statement	Loop until condition is true
RESIGNAL statement	Re-raise an exception
SIGNAL statement	Raise an exception
WHILE statement	Loop while condition is true

Auxiliary statements

Collect statistics, manage caching, explore metadata, set configurations, and manage resources.

ANALYZE statements

Statement	Description
ANALYZE TABLE … COMPUTE STATISTICS	Compute table column statistics
ANALYZE TABLE … COMPUTE STORAGE METRICS	Compute table storage metrics

Cache statements

Applies to: Databricks Runtime

Statement	Description
CACHE TABLE	Cache table in memory
CLEAR CACHE	Remove all cached data
REFRESH CACHE	Refresh cached data
REFRESH FUNCTION	Refresh function metadata
REFRESH TABLE	Refresh table metadata
UNCACHE TABLE	Remove table from cache

DESCRIBE statements

Statement	Description
DESCRIBE CATALOG	Show catalog details
DESCRIBE CONNECTION	Show connection details
DESCRIBE CREDENTIAL	Show credential details
DESCRIBE DATABASE	Show database details
DESCRIBE FUNCTION	Show function details
DESCRIBE EXTERNAL LOCATION	Show location details
DESCRIBE PROCEDURE	Show procedure details
DESCRIBE PROVIDER	Show provider details
DESCRIBE QUERY	Show query output schema
DESCRIBE RECIPIENT	Show recipient details
DESCRIBE SCHEMA	Show schema details
DESCRIBE SHARE	Show share details
DESCRIBE TABLE	Show table details
DESCRIBE VOLUME	Show volume details

SHOW statements

Statement	Description
LIST	List files in a path
SHOW ALL IN SHARE	Show all objects in share
SHOW CATALOGS	List catalogs
SHOW COLUMNS	List table columns
SHOW CONNECTIONS	List connections
SHOW CREATE TABLE	Show table DDL
SHOW CREDENTIALS	List credentials
SHOW DATABASES	List databases
SHOW FUNCTIONS	List functions
SHOW GROUPS	List groups
SHOW EXTERNAL LOCATIONS	List external locations
SHOW PARTITIONS	List table partitions
SHOW PROCEDURES	List procedures
SHOW PROVIDERS	List providers
SHOW RECIPIENTS	List recipients
SHOW SCHEMAS	List schemas
SHOW SHARES	List shares
SHOW SHARES IN PROVIDER	List shares from provider
SHOW TABLE EXTENDED	Show table information
SHOW TABLES	List tables
SHOW TABLES DROPPED	List dropped tables
SHOW TBLPROPERTIES	Show table properties
SHOW USERS	List users
SHOW VIEWS	List views
SHOW VOLUMES	List volumes

Configuration and variable management

Statement	Description
CALL	Call a stored procedure
EXECUTE IMMEDIATE	Execute dynamic SQL
RESET	Reset configuration to default
SET	Set configuration parameter
SET RECIPIENT	Set share recipient context
SET TIME ZONE	Set session timezone
SET variable	Set variable value
USE CATALOG	Set current catalog
USE DATABASE	Set current database
USE SCHEMA	Set current schema

Resource management

Applies to: Databricks Runtime

Statement	Description
ADD ARCHIVE	Add archive to session
ADD FILE	Add file to session
ADD JAR	Add JAR to session
LIST ARCHIVE	List session archives
LIST FILE	List session files
LIST JAR	List session JARs

Applies to: Databricks SQL Connector

Statement	Description
GET	Download file from volume
PUT INTO	Upload file to volume
REMOVE	Delete file from volume

Security statements

Manage access to data. For details, see Hive metastore privileges and securable objects (legacy).

Statement	Description
ALTER GROUP	Modify group membership
CREATE GROUP	Create a new group
DENY	Deny privileges
DROP GROUP	Delete a group
GRANT	Grant privileges
GRANT ON SHARE	Grant share access
MSCK REPAIR PRIVILEGES	Repair privilege metadata
REVOKE	Revoke privileges
REVOKE ON SHARE	Revoke share access
SHOW GRANTS	Show granted privileges
SHOW GRANTS ON SHARE	Show share privileges
SHOW GRANTS TO RECIPIENT	Show recipient privileges