IBM FileNet P8, Version 5.2.1            

Conversion rules for mapped property values

All mapped properties undergo two conversions: initially from native data type to string, and then from string to Content Platform Engine data type.

Content Integrator first converts all mapped property values regardless of their native data type to Unicode string values by using a set of conversion rules. The rules depend on the Content Integrator implementation for the connector that is used and on the property description.

The property values are reconverted by IBM® FileNet® CFS for Content Integrator to the data type of the corresponding Content Platform Engine property. When IBM FileNet CFS for Content Integrator assigns a value to a property on an object in the target object store, the value must conform to the fundamental data type that is specified by the property definition or the value must allow for implicit conversion to that data type without loss of information. For example, if the property definition on the target class specifies a type of integer, then the mapped property must also have integer values or a string that consists of numeric characters that can be implicitly converted to an integer value. A string value that includes non-numeric characters, however, results in a constraint violation because it cannot be implicitly converted to an integer value.

The following table specifies the conversion rules for supported data types.

Unicode to Content Platform Engine data type Conversion rule
String values to string values All properties are converted to strings as part of the mapping process. No conversion is required when IBM FileNet Content Federation Services for Content Integrator assigns a value to a Content Platform Engine property that is defined with a fundamental data type of string.
String values to Boolean values A Boolean value of TRUE is created from a string value of true when ignoring cases. Otherwise, the string value is converted to Boolean FALSE.
String values to signed long values The characters in the string value must all be decimal digits, except that the first character can be an ASCII minus sign '-' (\u002D') to indicate a negative value.
String values to DateTime values The characters in the string value must be decimal digits that represent the number of milliseconds since 00:00:00 GMT on January 1, 1970.
String values to double values The characters in the string must describe a double value according to the lexical structure detailed in The Java™ Language Specification.
String values to ID values The string must represent a GUID value. Valid GUID values can optionally have curly braces on either end, and must consist of 8 hexadecimal characters, a dash, 4 hexadecimal characters, a dash, 4 hexadecimal characters, a dash, 4 hexadecimal characters, a dash, and 12 hexadecimal characters. For example: {1234abcd-56ef-7a89-9fe8-7d65cd43ba21}.


Last updated: October 2015
p8pca048.htm

© Copyright IBM Corporation 2015.