This content has been marked as final. Show 12 replies
It depends which data you need to store in your db. If you are working wiht Chinese character than you need to choose AL32UTF8, it supports all langauges.
Thatswhy We prefer this set always.
Pelase see below note for more details
The National Character Set ( NLS_NCHAR_CHARACTERSET ) in Oracle 9i, 10g and 11g [ID 276914.1]
Refer to :http://www.oracle-base.com/articles/10g/CharacterSetMigration.php
- Pavan Kumar N
What is the disadvantage of using WE8ISO8859P1 ( in 11g WE8MSWIN1252 ) as the database character set and using AL16UTF16 as the National character set.
We had some chinese characters and I believe we need to use Nchar datatype
Please check below link
How to Change the Character Set from US7ASCII to UTF8 after Database Creation [ID 408100.1]
Difference between WE8MSWIN1252 and WE8ISO8859P1 characterset [ID 341676.1]
Choosing between WE8ISO8859P1, WE8ISO8859P15 or WE8MSWIN1252 as NLS_CHARACTERSET [ID 264294.1]
According to docs, in 11g WE8ISO8859P1 is replaced by WE8MSWIN1252.
Therefore I have to use WE8MSWIN1252.
Suppose I have some chinese characters.
What is the disadvantage of using WE8MSWIN1252 and use AL16UTF16 as the National character set ? I will use nvarchar2 dataype for chinese characters.
What is the advantage of not using AL32UTF8 as a database character set ?
There is no specific doc to compare those 2 char set.
Please check below notes:
Character Sets & Conversion - Frequently Asked Questions [ID 227330.1]
AL32UTF8 / UTF8 (Unicode) Database Character Set Implications [ID 788156.1]
These notes are not what I am looking for since I am not after a conversion.
If you know you're going to use chinese characters and you're starting from scratch so you have the option of choosing the right characterset up front, that's what I'd do.
NCHAR, NVARCHAR2, NCLOB is a complication you can avoid.
Downside - a multibyte characterset will use more space than a single byte characterset. Certainly not a big deal if a multibyte characterset is really what you need.
Bottom line is,
Downside of using national character set is complication (nchar,nvarchar2,etc) and downside of using unicode is additional storage.
Is that right ? Apart from these there are no disadvantages.
Then if you're going to use a multibyte characterset, you just need to remember about CHAR length semantics for the relevant columns.
Chapter 6 of the Globalization Support Guide compares the two options - http://download.oracle.com/docs/cd/E11882_01/server.112/e10729/ch6unicode.htm - see section titled "How to Choose Between a Unicode Database and a Unicode Data Type Solution"