Relativity supports many standard character sets. If you need to define a character set that is nonstandard, however, Relativity has the ability to map data stored in one character set to a different character set for display purposes. This is usually necessary when data that was initially stored using one character set standard must be displayed using another character set standard; for example, data stored on a UNIX server using an ISO 8859 standard character set that must be displayed on Windows using a Code Page standard character set. ISO 8859 and Code Page character set standards are similar, but not identical, in the 8-bit values that are used to represent the different display characters. When the standards disagree on the 8-bit values (known as Code Points), the actual character displayed (known as a Glyph) will be incorrect.
To describe a character set, Relativity uses a derivation of the Unicode Consortium's Format A character set definitions.
When the mapping tables are created between two character sets, unused Code Points and Code Points with unmatched Glyphs have a simple, one-to-one mapping between the two character sets. This mapping proceeds from the lowest Code Point to the highest and is symmetrical. That is, a character that is read and mapped through these mapping tables will have the correct value stored on a subsequent UPDATE operation.
If the data within a file does not match any of the character sets already defined in Relativity, then the character set of the data file must be installed in the Relativity data source. By so doing, Relativity can build the mapping tables later.