Oracle® Database System Administration Guide 10g Release 2 (10.2) for IBM z/OS (OS/390) Part Number B25398-01 |
|
|
View PDF |
This appendix documents the National Language Support (NLS) information specific to Oracle Database for z/OS. Information about the product-specific operation of language-specific features is provided in Oracle Database Globalization Support Guide.
The topics in this appendix include:
Oracle Database Globalization Support enables you to store, process, and retrieve data in native languages. National Language Support (NLS) is a subset of globalization support. It enables Oracle applications to interact with users in their native language, using their specific cultural conventions for displaying data.
The Oracle NLS architecture is data-driven, enabling support for specific languages and character encoding schemes to be added without requiring any changes in source code.
Oracle Database for z/OS currently supports 24 languages. This table lists the languages that are supported and the default territories for each.
Language | Default Territory |
---|---|
American | America |
Arabic | United Arab Emirates |
Bengali | Bangladesh |
Brazilian Portuguese | Brazil |
Bulgarian | Bulgaria |
Canadian French | Canada |
Catalan | Catalonia |
Croatian | Croatia |
Czech | Czechoslovakia |
Danish | Denmark |
Dutch | Netherlands |
Egyptian | Egypt |
English | United Kingdom |
Estonian | Estonia |
Finnish | Finland |
French | France |
German | Germany |
German DIN | Germany |
Greek | Greece |
Hebrew | Israel |
Hungarian | Hungary |
Icelandic | Iceland |
Indonesian | Indonesia |
Italian | Italy |
Japanese | Japan |
Korean | Korea |
Latin American Spanish | America |
Latvian | Latvia |
Lithuanian | Lithuania |
Malay | Malaysia |
Mexican Spanish | Mexico |
Norwegian | Norway |
Polish | Poland |
Portuguese | Portugal |
Romanian | Romania |
Russian | CIS |
Simplified Chinese | China |
Slovak | Czechoslovakia |
Slovenian | Slovenia |
Spanish | Spain |
Swedish | Sweden |
Thai | Thailand |
Traditional Chinese | Taiwan |
Turkish | Turkey |
Ukranian | Ukraine |
Vietnamese | Vietnam |
Oracle automatically converts these types of data as they transfer between client and server, if required:
CHAR, VARCHAR, and LONG database columns
SQL and PL/SQL statements
host variables containing character data
Both the client and the server have associated character sets. The client declares its character set before connecting to the server through the NLS_LANG environment variable. On z/OS, this parameter is in the ORA$ENV DD statement.
If NLS_LANG is not specified, the default character set is assigned.
The character set for the server is declared when a database is created in Oracle and it cannot be changed once established. The default database character set is assigned if one is not explicitly declared.
When the client character set matches the server character set, character data is sent between client and server without any conversion. If the two character sets differ, all character data is converted from one character set to the other as it is transferred. It is important to be aware that all data contained in a database, whether in user-specified tables or in Oracle-specified data dictionary tables, is stored in the database character set.
The character set in which the data is stored in the Oracle database is specified in the CHARACTER SET clause of the CREATE DATABASE statement. Refer to the Oracle Database Administrator's Guide for more information about the CREATE DATABASE statement.
When creating a database, the character set you choose depends on the language(s) to be supported and the character set(s) of the clients connecting to the server. Many languages can be supported by the default z/OS character set of WE8EBCDIC1047, but many others cannot.
If your site requires Euro support then you need to consider moving your server code page to WE8EBCDIC924. This code page is the EBCDIC version of ISO 8859-15.
The following table allows you to select an appropriate value for the CHARACTER SET clause of the CREATE DATABASE statement, based on the language to be supported and on the character set of the z/OS client. If you need to support languages that are not on this chart, please contact Oracle Support Services.
Language | Client Character Set on z/OS | Server Character Set on z/OS |
---|---|---|
Arabic | AR8EBCDICX | AR8EBCDIC420S |
Baltic | BLT8EBCDIC1112 | BLT8EBCDIC1112S |
Cyrillic | CL8EBCDIC1025 | CL8EBCDIC1025RFoot 1 |
Eastern European | EE8EBCDIC870 | EE8EBCDIC870S |
Greek | EL8EBCDIC875 | EL8EBCDIC875R1 |
Hebrew | IW8EBCDIC424 | IW8EBCDIC424S |
Icelandic | WE8EBCDIC871 | WE8EBCDIC871S |
Japanese | JA16DBCS | JA16DBCS |
Korean | KO16DBCS | KO16DBCS |
Simplified Chinese | ZHS16DBCS | ZHS16DBCS |
Thai | TH8TISEBCDIC | TH8TISEBCDICS |
Traditional Chinese | ZHT16DBCS | ZHT16DBCS |
Turkish | TR8EBCDIC1026 | TR8EBCDIC1026S |
Western European | WE8EBCDIC924 | WE8EBCDIC924 |
Western European | WE8EBCDIC1047 | WE8EBCDIC1047 |
Western European | WE8EBCDIC1047E | WE8EBCDIC1047E |
The client character set is determined by the data that is sent to the Oracle server. This is typically determined by the type of terminal that is used by the client. See your system administrator for more information about the character set used by your terminal.
The default client character set has changed from WE8EBCDIC37C to WE8EBCDIC1047, but most existing Oracle7 for OS/390 databases are WE8EBCDIC37C. When the client and server character sets differ, the CPU overhead increases. If you set NLS_LANG so that the client and server character sets match, the CPU overhead decreases.
The following character sets are supported for z/OS clients:
Character Set | Support |
---|---|
AR8EBCDICX | XBASIC Code Page 420 Arabic |
BLT8EBCDIC1112 | EBCDIC Code Page 1112 Baltic/Multilingual |
CL8EBCDIC1025 | EBCDIC Code Page 1025 Cyrillic/Multilingual |
D8EBCDIC273 | EBCDIC Code Page 273 Austrian /German |
DK8EBCDIC277 | EBCDIC Code Page 277 Danish |
EE8EBCDIC870 | EBCDIC Code Page 870 East European |
EL8EBCDIC875 | EBCDIC Code Page 875 Greek |
F8EBCDIC297 | EBCDIC Code Page 297 French |
I8EBCDIC280 | EBCDIC Code Page 280 Italian |
IW8EBCDIC424 | EBCDIC Code Page 424 Hebrew |
JA16DBCS | EBCDIC DBCS Japanese |
KO16DBCS | EBCDIC DBCS Korean |
S8EBCDIC278 | EBCDIC Code Page 278 Swedish |
TH8TISEBCDIC | EBCDIC Code Page 838 Thai |
TR8EBCDIC1026 | EBCDIC Code Page ISO 8859 Turkish |
WE8EBCDIC37 | EBCDIC Code Page 37 West European |
WE8EBCDIC37C | EBCDIC Code Page 37 West European with extensions |
WE8EBCDIC284 | EBCDIC Code Page 284 Spanish (Spain) |
WE8EBCDIC285 | EBCDIC Code Page 285 English, UK |
WE8EBCDIC500 | EBCDIC Code Page 500 West European |
WE8EBCDIC871 | EBCDIC Code Page 871 Icelandic |
WE8EBCDIC1047 | EBCDIC Code Page 1047 Latin 1/Open Systems |
WE8EBCDIC1140 | EBCDIC code page 1140 Western European with Euro support |
WE8EBCDIC1140C | EBCDIC code page 1140 Western European with extensions and Euro support |
ZHS16DBCS | EBCDIC DBCS Simplified Chinese |
ZHT16DBCS | EBCDIC DBCS Traditional Chinese |
The LXEGEN utility is used to customize calendar data, as described in Oracle Database Globalization Support Guide. Customizing calendar data is done primarily through z/OS UNIX System Services. You will need to be familiar with z/OS UNIX System Services in general, as well as the material in Oracle Database User's Guide for IBM z/OS describing the particular environment variables required by Oracle products in a z/OS UNIX System Services shell environment.
The z/OS implementation of Oracle products uses the same filenames and directory structure that the Oracle product documentation uses. Log on to z/OS UNIX System Services and verify that you can locate the applicable directories and files. If you have any difficulty locating these files, ensure that the ORACLE_HOME environment value is properly set and that z/OS UNIX System Services is properly installed before attempting to customize yor calendar data.