How to use the SQL reference

Applies to: Databricks SQL Databricks Runtime

This guide explains how to read and navigate the SQL language reference, including platform availability labels and syntax notation.

"Applies to" labels

The SQL reference covers Databricks SQL, Databricks Runtime, and Unity Catalog. Each article includes an "Applies to" label near the top that shows which products and versions support that feature.

Reading the label

Label	Meaning
Databricks SQL Databricks Runtime	All versions of both Databricks SQL and Databricks Runtime
Databricks SQL	Databricks SQL only (not supported in Databricks Runtime)
Databricks SQL Databricks Runtime 11.3 LTS and above	All versions of Databricks SQL and Databricks Runtime 11.3 LTS and later
Databricks SQL Databricks Runtime 11.3 LTS and above Unity Catalog only	Unity Catalog only, all versions of Databricks SQL, and Databricks Runtime 11.3 LTS and later

Parameter-level labels

Some articles have an "Applies to" label at the top and additional labels for specific parameters within the article. The parameter-level label overrides the article-level label for that parameter only.

Example 1: Parameter requires a minimum Databricks Runtime version

Article label: Databricks SQL Databricks Runtime

Parameter label: Databricks SQL Databricks Runtime 11.3 LTS and above

This means most of the article applies to all versions of both Databricks SQL and Databricks Runtime, but that specific parameter requires Databricks Runtime 11.3 LTS or later.

Example 2: Parameter is not available for SQL warehouse workloads

Article label: Databricks SQL Databricks Runtime

Parameter label: Databricks Runtime 11.3 LTS and above

This means most of the page applies to all versions of both products, but that specific parameter is only supported in Databricks Runtime 11.3 LTS and later (not supported in Databricks SQL).

---

Syntax diagrams

SQL statements are documented using syntax diagrams that show how to construct valid commands.

Basic elements

Syntax diagrams use these foundational components to represent SQL statements. Each component is followed by an example.

Keyword

Keywords appear in uppercase but are case-insensitive.

SELECT

Token

Literal characters that you must enter exactly as shown: (, ), <, >, ., *, ,

Clause

Clauses are named syntax sections. Uppercase clauses (like LIMIT clause) link to their documentation. Lowercase clauses with underscores (like named_expression) are defined locally.

LIMIT clause
named_expression

Argument

Function arguments use camelCase and are described in the Arguments section of Databricks SQL function reference.

mapExpr

Syntax notation

Special symbols indicate how components can be combined, repeated, or omitted.

Required sequence

Components separated by whitespace must appear in order.

SELECT expr

Mandatory choice

Curly braces { } with | separators mean you must choose exactly one option.

{ INT | INTEGER }

Optional choice

Square brackets [ ] with | separators mean you can choose at most one option.

[ ASC | DESC ]

Grouping

Curly braces { } without | mean you must include all components.

{ SELECT expr }

Optional element

Square brackets [ ] mean the enclosed component is optional.

[ NOT NULL ]

Repetition

Ellipsis [...] means you can repeat the preceding component. If preceded by a separator (, ...), you must use that separator between repetitions.

col_option [...]
col_alias [, ...]
{ expr [ AS ] col_alias } [, ...]

---

Comments

Databricks SQL supports C-style comments (/* ... */) and line comments (-- ...).