Contents: | Main | Chapter | See Also: | Getting Started Manual | Advanced User Manual |
You can make a data audit conditional when you define a field as being audited. An audit condition is a line of M code with the characteristics that follow:
Variable | Description | |
DA | Internal number of the entry being audited. The DA-array will exist if the audit is in a subfile. | |
DIE | The global root of the file or subfile being audited. | |
DIIX | A two-piece variable described below: | |
piece 1 | 3 if this audit is taking place during a set; and 2 if this audit is taking place during a kill. | |
piece 2 | Field number being edited. | |
X | The internal representation of a field's value, i.e., the actual stored value. X is always present, but its value will vary based on the first piece of DIIX. If $P(DIIX,U,1)=3, then X equals the new value in the field. If $P(DIIX,U,1)=2, then X equals the old value in the field. |
If the data type of the field being audited is a Pointer, Variable Pointer, or Set of Codes, then the internal value of the field and its data type will be stored. The old value is stored on node 2.1 of the entry in the Audit file (#1.1) and the new value is stored on node 3.1.
Reviewed/Updated: March 10, 2007