Glossary uuid

Version 7 of the Sage X3 engine introduced a new data type called UUID Universally Unique Identifier. This is a standard published by the Open Software Foundation (OSF).

The purpose of UUIDs is to allow unique identifiers to be assigned without coordination in distributed systems. There is no system that will guarantee that this identifier is unique because a UUID is a key with a finite length. The UUID generation process has been designed to be practically unique, and this is the case for database IDs. UUIDs can be used to identify a piece of information with a reasonable confidence that nobody will unintentionally identify another piece of information with the same UUID and thus create a conflict.

All operating systems provide a function to generate UUIDs. The algorithm usually combines the current time, a unique number assigned to the computer, and a random number.

A UUID is stored on 128 bits (16 bytes). A canonical printable form represents them as 32 hexadecimal digits, displayed in five groups separated by hyphens. The format is 8-4-4-4-12, which totals 36 characters including 32 digits and 4 hyphens (for example, 02311fc1-8bf6-4b90-814d-4b456b03faf2). There are approximately 3*10^38 possible UUIDs.

The Sage X3 engine includes a set of instructions and functions to support UUIDs:

• Uuident: Use to declare a UUID variable
• Nulluuid: Use to manage a null (not assigned) UUID
• Touuid: Use to transform a canonical form to a UUID
• Getuuid: Use to generate a UUID

The supervisor layers, especially CRUD operation support and input control operations, use UUIDs intensively. For example:

• By default, a UUID column is now present on every Sage X3 table.
• When grids are managed by Sage X3 clients, lines are identified by UUIDs. This makes it very easy to identify array lines in insertion, deletion, and sort operations.