Using SQLite

Appendix C. SQLite SQL Command Reference

This appendix lists the SQL commands and syntax that are supported by SQLite. SQL statements consist of a single command and any required parameters. Command statements are separated by a semicolon. Technically, standalone statements do not need to be terminated with a semicolon, but most interactive environments require the use of a semicolon to indicate that the current command statement is complete and should be executed. For example, the C API sqlite3_exec() does not require that command statements end with a semicolon, but interactive use of sqlite3 requires ending each statement with a semicolon.

In most situations where a table name is called for, a view name can be used instead. As noted in the syntax diagrams, in most instances where any object identifier is used (table name, view name, etc.), the name can be qualified with a logical database name to prevent any ambiguity between objects in different databases that share a similar name (see ATTACH DATABASE in this appendix). If the object is unqualified, it will be searched for in the temp database, followed by the main database, followed by each attached database, in order. If an unqualified identifier appears in a CREATE statement, the object will be created in the main database, unless the statement contains some type of CREATE TEMPORARY syntax. Object identifiers that use nonstandard characters must be quoted. See Basic Syntax for more info.

The SELECT, UPDATE, and DELETE commands ...