Character set considerations

Each character of text is represented in one or more bytes. The mapping from characters to binary codes is called the character set encoding. Some character sets used for languages with small alphabets, such as European languages, use a single-byte representation. Others, such as Unicode, use a double-byte representation. Because they use twice the storage space for each character, double-byte character sets can represent a much larger number of characters.

Conversion errors can occur or data can be lost when text using one character set must be converted to another character set. Not all characters can be represented in all character sets. In particular, single-byte character sets can represent a much smaller number of characters than multibyte systems because of the limited number of codes available.

When the character set of your MobiLink remote database is the same as your consolidated database, character conversion issues are avoided.

Text often needs to be sorted to build indexes and to prepare ordered result sets, such as directory listings. The sort order identifies the order of the characters. For example, a sort order typically states that the letter "a" comes before the letter "b", which comes before the letter "c".

Each database has a collation sequence. You set the collation sequence when you create the database, although how you do so can differ between database systems. The collation sequence defines both the character set and the sort order for that database.

Note

Whenever possible, define the collation sequence of your remote database to be the same as that of your consolidated database. This arrangement reduces the chance of erroneous conversions.

 See also

Character set conversion during synchronization