Data Movement Utilities Guide and Reference

Character Set and National Language Support

The DB2 UDB data movement utilities offer the following National Language Support (NLS):

• The import and the export utilities provide automatic code page conversion from a client code page to the server code page.
• For the load and the AutoLoader utilities, data can be converted from any code page to the server code page by using the codepage modifier with DEL and ASC files.
• For all utilities, IXF data is automatically converted from its original code page (as stored in the IXF file) to the server code page.

Unequal code page situations, involving expansion or contraction of the character data, can sometimes occur. For example, Japanese or Traditional-Chinese Extended UNIX Code (EUC) and double-byte character sets (DBCS) may encode different lengths for the same character. Normally, comparison of input data length to target column length is performed before reading in any data. If the input length is greater than the target length, NULLs are inserted into that column if it is nullable. Otherwise, the request is rejected. If the nochecklengths modifier (see File Type Modifiers (Load)) is specified, no initial comparison is performed, and an attempt is made to load the data. If the data is too long after translation is complete, the row is rejected. Otherwise, the data is loaded.