U-SQL Syntax Conventions (U-SQL)
Updated: March 10, 2017
This document uses an Extended Backus-Naur syntax notation of the following structure:
Each syntax grammar rule has a name followed by the rule body, as in:
name := rule.
Each rule body is either a reference to a name or a composition of the following rule expressions recursively. The following table lists and describes conventions that are used in the syntax diagrams in the U-SQL Language Reference.
| Convention | Used for |
|---|---|
| UPPERCASE | U-SQL keywords. |
[rule] | Brakets indicate that the grammar rule rule occurs either 0 or 1 times. Do not type the brackets. |
{rule} | Curly braces indicate that the grammar rule rule occurs 0 to n times. Do not type the braces. |
(rule) | Parenthesis are used to nest and group grammar rules. Do not type the parenthesis. |
rule1 | rule2 | The vertical line represents a choice between rule1 or rule2. It binds weaker than any of the other grammar rules. You can use only one of the items. |
'abc' | Literal keyword abc. |
'abc'+rule | Literal prefix abc tied to result of grammar rule rule without optional whitespace. |
| ; | U-SQL statement terminator. |
Upper-case rule names inside a rule definition are defined in the language reference documentation. Lower-case rule names are considered self-evident and not further specified.
See Also
Community Additions
ADD
Show: