Field help (ADSCODDBA)

Defines the set of characters used to store the character fields in the database. It can take the values ASCIIor UNICODE.

• The ASCII format is used for managing European languages: each character is stored on one byte and the accented characters on values higher than 128.

• The UNICODE format is only useful when managing languages having sets of characters requiring over 256 combinations. This is the case for Chinese, for example.
  The SAFE X3 client is native UNICODE: it inherently ‘knows’ how to manage texts of this type. It can elect to choose one of the two following formats at the level of the database: UCS2 or UTF8.
  
  The UCS2 format is a format originating with Microsoft™ in which all characters are stocked in 2 bytes; the normal ASCII characters have the same code (but one of the two bytes in which they are coded is null). This is the only UNICODE format type that is supported by SQLServer. On the other hand Oracle accepts other formats, and in particular the UTF8 format, which is the most common. It features a format in which the characters are stored over a variable number of bytes (from 1 to 5 depending on the cases, the standard ASCII characters are stored over one byte, the accented characters use two and the Asian languages start beyond). It should be noted that Oracle supports other coding standards (UCS3, UCS4, UTF16...) and these standards can be used (on the condition that they are created manually in the database).

Internally and independently of the database format (for its temporary variables), the SAFE X3 engine uses the UTF8 format (the process sources are coded in UTF8), and the Windows client uses the UCS2 standard.