NLS considerations

IBM i Access for Web displays information from a variety of sources. Some of these sources are able to provide information in more than one language, or format information in a language-specific manner.

These sources include:

  • IBM i
  • JDBC driver
  • Web application server
  • IBM i Access for Web

Such language-specific information includes forms, error messages, help, formatted dates and times, and sorted lists. When a choice of languages is available, every attempt is made to select a language that is appropriate for the user. The selected language is used to influence the language and format of information from these other sources. However, there is no guarantee that the information from a specific source will be in the selected language, or that all the information will be in the same language.

Messages and help which originate from the IBM i might not be in the selected language, depending on which language versions are installed on the IBM i. If the selected language is not installed on the IBM i, then IBM i messages displayed by IBM i Access for Web are in the primary language of the IBM i.

Language and character set selection

IBM i Access for Web uses the following method to select an appropriate language and character set.

First, a list of potential language choices is assembled from the following sources:

  • The IBM i Access for Web locale= parameter.
  • The IBM i Access for Web Preferred language preference.
  • The browser language configuration (HTTP Accept-Language header).
  • The IBM i user profile Language ID.
  • The Java™ Virtual Machine default locale.

Second, a list of acceptable character sets is assembled from the following sources:

  • The IBM i Access for Web charset= parameter.
  • The IBM i Access for Web Preferred character set preference.
  • The browser character set information (HTTP Accept-Charset header).
    Note: If the browser character set information (HTTP Accept-Charset header) indicates that UTF-8 is supported, the list of acceptable character sets will include UTF-8 before other character sets supported by the browser. In this case, UTF-8 is used as the character set when the Preferred character set preference is set to None.

Third, each language in the list of potential languages is examined to determine if it is available and if it can be represented using a character set from the list of acceptable character sets.

The first language that is available and can be represented using one of the acceptable character sets is selected.

The first character set from the list of acceptable character sets that can represent the language is selected.

Information in multiple languages (multilingual)

Because the information displayed by IBM i Access for Web comes from a variety of sources, there is a possibility that the information is in more than one language. When multiple languages are displayed in a browser simultaneously, a multilingual character set, such as UTF-8, might be required to display all the characters correctly. If this is the case, the Preferred character set preference should be changed to either Multilingual [UTF-8] or None.
Note: Since most browsers support UTF-8 as a character set, if you specify None for the Preferred character set preference, IBM i Access for Web uses UTF-8 for the character set.

CCSIDs and IBM i messages

To ensure that information is displayed properly, make certain that the Coded Character Set ID (CCSID) setting for the user profile is appropriate for the messages originating from IBM i.