Oracle Data Structures - Character Datatypes (Page 2 of 4 )
Character datatypes can store any string value, including the string representations of numeric values. Assigning a value larger than the length specified or allowed for a character datatype results in a runtime error. You can use string functions, such as UPPER, LOWER, SUBSTR, and SOUNDEX, on standard (not large) character value types.
There are several different character datatypes:
CHAR
The CHAR datatype stores character values with a fixed length. A CHAR datatype can have between 1 and 2,000 characters. If you don’t explicitly specify a length for a CHAR, it assumes the default length of 1. If you assign a value that’s shorter than the length specified for the CHAR datatype, Oracle will automatically pad the value with blanks. Some examples of CHAR values are: CHAR(10) = "Rick | ", "Jon | ", "Stackowiak" |
VARCHAR2
The VARCHAR2 datatype stores variable-length character strings. Although you must assign a length to a VARCHAR2 datatype, this length is the maximum length for a value rather than the required length. Values assigned to a VARCHAR2 datatype aren’t padded with blanks. The VARCHAR2 datatype can have up to 4,000 characters. Because of this, a VARCHAR2 datatype can require less storage space than a CHAR datatype, because the VARCHAR2 datatype stores only the characters assigned to the column.
At this time, the VARCHAR and VARCHAR2 datatypes are synonymous in Oracle8 and later versions, but Oracle recommends the use of VARCHAR2 because future changes may cause VARCHAR and VARCHAR2 to diverge. The values shown earlier for the CHAR values, if entered as VARCHAR2 values, are:
VARCHAR2(10) = "Rick", "Jon", "Stackowiak"
NCHAR and NVARCHAR2
The NCHAR and NVARCHAR2 datatypes store fixed-length or variable-length character data, respectively, using a different character set from the one used by the rest of the database. When you create a database, you specify the character set that will be used for encoding the various characters stored in the database. You can optionally specify a secondary character set as well (which is known as the National Language Set, or NLS). The secondary character set will be used for NCHAR and NVARCHAR2 columns. For example, you may have a description field in which you want to store Japanese characters while the rest of the database uses English encoding. You would specify a secondary character set that supports Japanese characters when you create the database, and then use the NCHAR or NVARCHAR2 datatype for the columns in question.
Starting with Oracle9i, you can specify the length of NCHAR and NVARCHAR2 columns in characters, rather than in bytes. For example, you can indicate that a column with one of these datatypes is 7 characters. The Oracle9i database will automatically make the conversion to 14 bytes of storage if the character set requires double-byte storage.
Oracle Database 10g introduced the Globalization Development Kit (GDK), which is designed to aid in the creation of Internet applications that will be used with different languages. The key feature of this kit is a framework that implements best practices for globalization for Java and PL/SQL developers.
Oracle Database 10g also added support for case-and accent-insensitive queries and sorts. You can use this feature if you want to use only base letters or base letters and accents in a query or sort.
LONG
The LONG datatype can hold up to 2 GB of character data. It is regarded as a legacy datatype from earlier versions of Oracle. If you want to store large amounts of character data, Oracle now recommends that you use the CLOB and NCLOB datatypes. There are many restrictions on the use of LONG datatypes in a table and within SQL statements, such as the fact that you cannot use LONGs in WHERE, GROUP BY, ORDER BY, or CONNECT BY clauses or in SQL statements with the DISTINCT qualifier. You also cannot create an index on a LONG column.
CLOB and NCLOB
The CLOB and NCLOB datatypes can store up to 4 GB of character data prior to Oracle Database 10g. Starting with Oracle Database 10g, the limit has been increased to 128 TBs, depending on the block size of the database. The NCLOB datatype stores the NLS data. Oracle Database 10g and later releases implicitly perform conversions between CLOBs and NCLOBs. For more information on CLOBs and NCLOBs, please refer to the discussion about large objects (LOBs) in the section “Other Datatypes ” later in this chapter.
Next: Numeric Datatype >>
More Database Articles Articles
More By O'Reilly Media
|
 This article is excerpted from chapter four of the book Oracle Essentials, Fourth Edition Oracle Database 11g, written by Rick Greenwald, Robert Stackowiak, and Jonathan Stern (O'Reilly, 2007; ISBN: 0596514549). Check it out today at your favorite bookstore. Buy this book now.
|
|
