GAS System Encoding

System character encoding matters when Genero Application Server (GAS) interacts with the operating system.

For example, when the GAS:

Writes log files.
Opens files defined in the GAS configuration file (as.xcf).
Reads arguments from the command line.

In these cases and more, GAS uses the character encoding set by the operating system:

Linux^®/UNIX™: encoding is defined via environment variables LANG or LC_ALL. For more information see the "Localization" topics in Genero Business Development Language User Guide or see The Single UNIX - Specification Version 2 - Locale.
Windows^®: GAS defaults to the system locale as defined in the language and regional settings.
Note: There should be no need to set the LANG variable, except your application uses a different character set to the Windows system locale.

How GAS does character set conversion

The GAS software takes care of character set conversions:

For xcf files, it does the conversion based on what the XML prolog specifies as charset to the GAS locale, for example:
<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
For the front-ends such as Genero Web Client for JavaScript (GWC-JS) that use UTF-8 encoding:
- For template files, the charset in the metadata of $FGLASDIR/web/gwc-js/bootstrap.html file is used.
- For the DVM (V3) the conversion is done in the DVM locale, see DVM Locale.

Note: Operating system character sets may have different names across operating systems. To unify character set names in the application server environment, the GAS manages a character set encoding name conversion to map the operating system character encoding name to a canonical name:

A charset.alias file is provided. This file is located in the $FGLASDIR/etc directory of the GAS.