stewartc-708166 (4/29/2010)
Hugo Kornelis (4/29/2010)
I got it right, because of elimination, but I don't understand the entire scenario.Since the destination column is NVARCHAR(26), how can a value be outside the codepage range? I though codepages were relevant for non-Unicode data only?
in the situation that gave rise to this question, CODEPOINT identified the value of the character as 65533.
As far as I am aware, there is no charater (in any character set, albeit unicode or non-unicode) with that value.
This is very interesting. I am wondering where to practically draw the line with regard to how high a number resulting from a CODEPOINT test to draw the line between alpha/numeric characters from any language and other symbols/characters. Looking at this reference:
http://www.ssec.wisc.edu/~tomw/java/unicode.html I see there is a Unicode character(?) for the value 65533. Additional searching shows that Unicode goes up to 10FFFF or 1114111. I suspect the place to draw the line is either 65518 or 65276. I would certainly appreciate a link reference or a bit of feedback with regard to how to slice Unicode for the data flow in this question.
Thanks.