Working with cross-platform and localized data sources

If you are creating templates and data sources on operating systems other than those that will be serving or testing the template file, then you should specify the encoding system Generator uses when reading your data sources. If you don't, Generator may not interpret the data correctly.

For example, if you create a data source on a system with Japanese SJIS character codes and then serve that data source from a system with an English operating system, Generator will read each byte as a character instead of using two bytes for each character, thereby producing undesirable results.

Data sources created on Macintosh systems require MacRoman encoding when read by Windows or UNIX systems. Windows and UNIX systems commonly use ASCII encoding, but they can be altered to use other character encoding systems through operating system modification.

Generator supports encoding systems that cover U.S. English, Japanese, and most western European single-byte character encoding systems. It does not support fonts or foreign language characters that are not implemented using one of these standard character encoding systems.

Data source encoding specifies the encoding system to be used when reading all data sources referenced in the template file. Therefore, when you plan and create data sources for a template, you should use the same character encoding system for all the data sources.

To specify the data source encoding for all data sources referenced in the template file:

1 Choose File > Publish Settings.
2 Select Generator in the Formats tab.
3 Click the Generator tab.
4 Choose an encoding method from the Data Encoding option. Default uses the encoding method of the system from which the template is served.