Re: Storage requirements of NCHAR columns

Ross schreef:

> The Oracle docs clearly suggest (in several places) that NCHAR columns
> are defined in terms of the number of code units. If this is true (and
> I suspect it isn't), an NCHAR(30) could only store 10 three-byte
> characters.

Where? I can find (Globalization Manual, Ch 6, on the benefits of UTF8):
You can specify the length of SQL CHAR types in number of UCS-2 code points.

CAN, not: you specify, or you do specify.

Also: UCS-2 is fixed, 2bytes. Each character is 2byte (aka 1 code point).
Also: the *default* national character set (referring NCHAR) is AL16UTF16, so it just might be the documentation writers took that for granted

-- 
Regards,
Frank van Bortel

Top-posting is one way to shut me up...