Accentuation of indigenous names in the Oracle database

Question

We have a situation of inclusion of records of indigenous names in the Oracle bank, where there is a variation of accentuation, and may have "e" or "u" with tilde. Our Oracle parameters (version 9i) are NLS_LANGUAGE = BRAZILIAN PORTUGUESE NLS_TERRITORY = BRAZIL

Answer 1

There are two distinct elements involved:

• The Charsets used to create your database.
• The charset used for data presentation.

When creating the database are defined NLS_CHARACTERSET and NLS_NCHAR_CHARACTERSET. These parameters can be checked by the query:

select * 
from nls_database_parameters 
where parameter='NLS_CHARACTERSET' 
or parameter='NLS_NCHAR_CHARACTERSET';

Regarding the charset used for data presentation, I suggest setting NLS_LANG, which has the format <LANGUAGE>_<TERRITORY>.<CHARACTER_SET>. In case we would have <LANGUAGE> = BRAZILIAN PORTUGUESE, <TERRITORY> = BRAZIL and <CHARACTER SET> which may be:

AL16UTF16
AL32UTF8
UTF8
US7ASCII      [X]
WE8ISO8859P1  [X]
WE8ISO8859P15 [X]
WE8MSWIN1252  [X]

The Charsets marked with [X] do not allow letters and or u with til. See for example the WE8ISO8859P1 (ISO 8859-1):

Therefore, you will need to use Unicode, for example:

NLS_LANG = BRAZILIAN PORTUGUESE_BRAZIL.UTF8