Consider a string containing a half-square Katakana and a Kanji character ア 山 (the half-square Katakana 'a' followed by the Kanji 'yama' for mountain) stored in a column declared as CHAR(3) CHARACTER SET KANJISJIS. The VantageCloud Lake representation of the string is:
B1 | 8E | 52 |
ア | 山 |
If the column is exported from a UTF8 session, the export width rule is n (n character in the column produces n bytes as output), the following data is exported (observe that ア expanded to three bytes, so truncation at the logical character boundary occurred between ア and 山 so no padding is required).
EF | BD | B1 |
ア |