Consider the following table:
CREATE TABLE ctable1 (c1 VARCHAR(11) CHARACTER SET KANJI1);
The following table shows the difference between SUBSTR and SUBSTRING in Teradata mode for KANJI1 strings from KanjiEBCDIC client character set, assuming c1 contains MN<ABC >P.
| Query | Return Value |
|---|---|
| SELECT SUBSTR(c1,2) FROM ctable1; | N<ABC>P |
| SELECT SUBSTR(c1,3,8) FROM ctable1; | <ABC> |
| SELECT SUBSTR(c1,4) FROM ctable1; | ABC >P The client application might not be able to properly interpret the resulting multibyte characters because the shift out (<) is missing.
|
| SELECT SUBSTRING(c1 FROM 2) FROM ctable1; |
N<ABC>P |
| SELECT SUBSTRING(c1 FROM 3 FOR 8) FROM ctable1; |
<ABC>P |
| SELECT SUBSTRING(c1 FROM 4) FROM ctable1; |
<BC>P |