10.2.0.2 Ent Ed, AIX5L
We are planning a new database and I'm having a little confusion on
what character set to use. We are not sure if in the future we may
need some double byte characters. For the most part we can suffice
with AMERICAN_AMERICA.WE8MSWIN1252 but it is possible that in the
future we need multibyte characters. I have been reading the
Globalization guide and the section in Tom Kyte's Expert Oracle
Database Architecture on this but I could be totally mis interrepting
something here. I haven't read chapter 6 on unicode in the
Globalization Guide (yet). I don't want to use multibyte data unless
it is necessary but I do want to have the option available.
- Am I correct in thinking that if we use NLS_CHARACTERSET =
WE8MSWIN1252 and NLS_NCHAR_CHARACTERSET = AL16UTF16 we shouldn't have
any issues with losing multibyte characters?
- On my windows client I go to a command prompt and issue the
following c:\> echo %NLS_LANG%,
I get %NLS_LANG% returned back to me. I'm assuming that this means I
don't have the variable set. How does my client then treat the data
that I am getting from the database?
- When I query v$nls_parameters does this show me what character set
my client session is using?
- When I query nls_database_parameters, does this show me what the
database has been setup as on the server side?
- Within sqlplus what does 'show parameter nls' tell me? I'm assuming
the init.ora nls settings.
Received on Wed Aug 22 2007 - 07:55:54 CDT
