• Chapter 1: Getting Started with Stylus Studio®
• Chapter 2: Editing and Querying XML
• Chapter 3: Converting Non-XML Files to XML
  • Overview of Convert to XML
  • Choosing an Input File
  • The Convert to XML Editor
  • Parts of an Input File
  • Working with Regions
  • Working with Fields
  • Controlling XML Output
  • Creating an Adapter
  • Using Adapters in Stylus Studio
  • More About Converting EDI
    • Supported EDI Dialects
    • Understanding Separator Characters
      • How Stylus Studio Sets Separator Characters
      • Invalid Separator Characters
      • Setting Processing Instruction
      • Syntax
      • Ways to Specify Control Characters
    • Using the Built-In EDI Adapter to Convert EDI to XML
    • Using Convert to XML to Convert EDI to XML
    • Validating XML from/to EDI
  • User-Defined Adapters Properties Reference
• Chapter 4: Working with XSLT
• Chapter 5: Creating XSLT Using the XSLT Mapper
• Chapter 6: Debugging Stylesheets
• Chapter 7: Defining XML Schemas
• Chapter 8: Defining Document Type Definitions
• Chapter 9: Writing XPath Expressions
• Chapter 10: Working with XQuery in Stylus Studio
• Chapter 11: Composing Web Service Calls
• Chapter 12: Accessing Relational Data as XML
• Chapter 13: Building XML Pipelines
• Chapter 14: Publishing XML Data
• Chapter 15: Integrating with Third-Party File Systems
• Chapter 16: Extending Stylus Studio
• Chapter 17: The Stylus Studio Java API
• Chapter 18: Stylus Studio GUI Reference

Home > Online Product Documentation > Table of Contents > Understanding Separator Characters

Understanding Separator Characters

EDI files consist of tokens separated by special characters. Separators typically include the following:

Values for some of these separators are set explicitly in the EDI data stream. For example, in X12, the component separator is the value of the ISA16 field - the "I15: Component Element Separator." In EDIFACT, separator values can be set using the contents of the UNA segment.

How Stylus Studio Sets Separator Characters

When writing XML to EDI, Stylus Studio overwrites separator values using a set of processing instructions (PIs). As long as the character has not been emitted to the output stream, it can be changed. For example, the segment terminator can be changed using the PI only before the first segment is written.

Invalid Separator Characters

You cannot use any of the following characters as EDI separators:

Setting Processing Instruction

Stylus Studio uses the following naming convention and default values for XML processing instructions used to set values for separator characters:

If a PI is used to set the value for one separator (say, A) to a default value being used by another separator (B), a new value will be chosen for the B separator unless one has already been specified.

Syntax

The syntax of the PI is just "<?" followed by the name, followed by a space, and then the new special character before a closing "?>" tag. Consider a few examples:

<?edi_segment \n?>
               
<?edi_release \\?>
               
<?edi_element $?>
               
<?edi_component \x0001?>
               
<?edi_decimal .?>
               

            

Here is another example, showing PIs used in context. In this example, an X12 document needed to be written to meet the following requirements: the segment terminator needed to be a carriage return; the element separator needed to be an asterisk; and the component separator needed to be a greater-than symbol. The two PIs and the Component Element Separator are shown in bold for clarity:

<?xml version="1.0" encoding="utf-8"?>
               
<?edi_segment \r?>
               
<?edi_element *?>
               
<X12>
               
	<ISA>
               
		<ISA01><!--I01: Authorization Information Qualifier-->00</ISA01>
               
		<ISA02><!--I02: Authorization Information-->          </ISA02>
               
		<ISA03><!--I03: Security Information Qualifier-->00</ISA03>
               
		<ISA04><!--I04: Security Information-->          </ISA04>
               
		<ISA05><!--I05: Interchange ID Qualifier-->01</ISA05>
               
		<ISA06><!--I06: Interchange Sender ID-->1515151515     </ISA06>
               
		<ISA07><!--I05: Interchange ID Qualifier-->01</ISA07>
               
		<ISA08><!--I07: Interchange Receiver ID-->5151515151     </ISA08>
               
		<ISA09><!--I08: Interchange Date-->041201<!--2004-12-01--></ISA09>
               
		<ISA10><!--I09: Interchange Time-->1217</ISA10>
               
		<ISA11><!--I65: Repetition Separator-->U</ISA11>
               
		<ISA12><!--I11: Interchange Control Version-->00403</ISA12>
               
		<ISA13><!--I12: Interchange Control Number-->000032123</ISA13>
               
		<ISA14><!--I13: Acknowledgment Requested-->0</ISA14>
               
		<ISA15><!--I14: Usage Indicator-->P</ISA15>
               
		<ISA16><!--I15: Component Element Separator-->></ISA16>
               
	</ISA>
               
...
               

            

If the segment terminator (edi_segment) is set to \r or \n, consider disabling the default value (specified in the URI) of adding end-of-line characters to each segment. Otherwise, the EDI output will have double endings. To do this, change eol=yes to eol=no in the URI. For more information, see Adapter Properties.

Ways to Specify Control Characters

In addition to literal characters, you can also specify control characters using a backslash syntax, as shown here.