Character Data vs Binary Data Considerations

If your application uses the COBOL wizard to create an ePortal server message, then all fields in this server message are always treated as strings.

If your application uses the X_OCTET wizard to create an ePortal server message, then the server message consists of a single unstructured blob of data. You can determine whether you want ePortal to treat this blob of data as a character string or as an array of binary bytes.

If your application uses the View wizard to create an ePortal server message, then the server message consists of one or more fields. Each field can have a different type. Like an X_OCTET buffer, the View field types char and carray can be used to represent either character data or binary data.

For each field of type char or carray, you can specify whether you want ePortal to treat the field as a character string or as an array of binary bytes.

In cases where data can be treated as either a string or an array of binary bytes, ePortal treats this data as a character string by default. If you want ePortal to treat this data as binary bytes, you need to edit the DTP record (with an extension of .dtpxml). Refer to the section Configuring String vs Binary Treatment in Creating an OS 2200 Open DTP Data Source Project.

Data Conversions

When ePortal sends a field containing binary bytes to the OS 2200, ePortal does not perform any conversion. Similarly, when ePortal receives a field containing binary bytes from the OS 2200, ePortal does not perform any conversion.

ePortal represents all character strings in Unicode. When ePortal sends a field containing a character string to the OS 2200, ePortal converts the string from Unicode to an OS 2200 character set that you can configure. When ePortal receives a field containing a character string from the OS 2200, ePortal converts the string from the OS 2200 character set back to Unicode.

To configure the OS 2200 character set, refer to Viewing or Changing ePortal Application Configuration Properties and ePortal Application Configuration Property Values.

Handling Multiple Character Encodings in a Single ePortal Message

Multi-byte or double-byte COBOL record definitions can contain data items that use different character encodings. The USAGE clause associated with a data item provides a hint as to which character encoding should be used to process the item.

The USAGE clause of DISPLAY is used for data items that contain multi-byte values. The configured OS 2200 character set described above is used to process this type of item. In addition, record definitions can contain a USAGE clause value of DISPLAY-2 to indicate the data item contains double-byte characters.

To handle data items that contain double-byte characters, ClearPath ePortal provides a mapping to a double-byte character encoding that is based on the configured OS 2200 character set. The following table shows that mapping.

Support for Dynamic Character Mapping

If you configure a multi-byte OS 2200 character set, you can optionally define a dynamic character map. Refer to Dynamic Character Mapping for more information.