> [] On Behalf Of Kyotaro
> Thanks, by the way, there's another issue related to SJIS conversion.  MS932
> has several characters that have multiple code points. By converting texts
> in this encoding to and from Unicode causes a round-trop problem. For
> example,
> 8754(ROMAN NUMERICAL I in NEC specials)
>   => U+2160(ROMAN NUMERICAL I)
>     => FA4A (ROMAN NUMERICA I in IBM extension)
> My counting said that 398 characters are affected by this kind of replacement.
> Addition to that, "GAIJI" (Private usage area) is not allowed. Is this meet
> your purpose?

Supporting GAIJI is not a requirement as far as I know.  Thank you for sharing 

# I realize my lack of knowledge about character sets...

Takayuki Tsunakawa

Sent via pgsql-hackers mailing list (
To make changes to your subscription:

Reply via email to