Delimiter Preservation and Suppression

  • Article

Overview

There are two properties that apply to delimited records: Preserve Delimiter For Empty Data and Suppress Trailing Delimiters. Use these properties to control how the flat file assembler handles delimiters associated with nonexistent data and trailing delimiters. When you set the Preserve Delimiter For Empty Data property to Yes (which is the default setting), delimiters are included in the translated flat file message for:

  • Fields without data.

  • Immediately subordinate records without data that do not have a tag associated with them.

    When you set the Preserve Delimiter For Empty Data property to No, delimiters are not included in the translated flat file for records and fields without data. Further, regardless of the setting of the Preserve Delimiter For Empty Data property, delimiters will not be included in the translated flat file message for immediately subordinate records without data for which a tag is defined.

    When you set the Suppress Trailing Delimiters property to No (which is the default setting), one or more trailing delimiters may be included in the translated flat file message. When you set the Suppress Trailing Delimiters property to Yes, trailing delimiters are not included in the translated flat file message.

Special scenarios

There are some special cases where the behaviors caused by the settings of the Preserve Delimiter For Empty Data and Suppress Trailing Delimiters properties can conflict. In such cases, the behaviors associated with the latter property, Suppress Trailing Delimiters, will take precedence. Further, there are some special cases where you will be warned about potential conflicts between the settings you have chosen for these two properties.

For example, consider a Record node defined with the following property values:

  • Node Name is MyRec

  • Tag Identifier is Rec

  • Child Delimiter is ,

  • Child Order is Infix

    And defined to contain five Field Element nodes with the following names (they could also be Field Attribute nodes or subordinate Record nodes):

  • FieldElem1

  • FieldElem2

  • FieldElem3

  • FieldElem4

  • FieldElem5

    Next, assume that the following mainly empty XML fragment, representing this Record node, is passed to the flat file assembler.

<MyRec>  
    <FieldElem1 />  
    <FieldElem2 />  
    <FieldElem3>Val</FieldElem3>  
    <FieldElem4 />  
    <FieldElem5 />  
</MyRec>

The following table shows the output produced, and the associated additional property setting requirements for the relevant schema nodes, based on different settings for the Preserve Delimiter For Empty Data (PDFED) and Suppress Trailing Delimiters (STD) properties.

PDFED setting STD setting Output Additional node requirements
Yes No Rec,,,Val,, None.
No Yes Rec,Val All Field Element nodes must be configured as optional.
Yes Yes Rec,,,Val Nodes named FieldElem4 and FieldElem5 must be configured as optional.
No No Rec,Val,, All Field Element nodes must be configured as optional.

When these property settings specify that delimiters should either not be preserved or should be suppressed, a message warning that it might not be possible to parse the serialized flat file data using the same schema will be issued in the following two cases:

  • When the Record node for which the Preserve Delimiter For Empty Data property is set to No and/or the Suppress Trailing Delimiters property is set to Yes, respectively, contains subordinate Field Element nodes, Field Attribute nodes, or Record nodes for which no tag is specified.

  • When the subordinate Field Element nodes, Field Attribute nodes, and Record nodes for which no tag is specified are not configured to be optional (by setting the Min Occurs property set to zero) in the schema. When the Suppress Trailing Delimiters property is set to Yes, only the last such subordinate nodes need to be configured as optional. When the Preserve Delimiter For Empty Data property is set to No, all trailing subordinate nodes need to be configured as optional.

Note

Delimiters are always preserved when the XML element associated with a (presumably optional) Record, Field Element, or Field Attribute node are entirely missing from the XML representation of the business document except when a Record follows a missing optional Field. In other words, when the data and its surrounding XML tags are both missing, the corresponding delimiter is always included in the flat file representation of the business document.

Now change the schema to include a child record with two Field Element immediately following a missing Field Element. The child record elements are configured to use the | (pipe) character as a delimiter.

<MyRec>  
    <FieldElem1 />  
    <FieldElem2 />  
    <FieldElem3>Val</FieldElem3>  
    <!-- <FieldElem4 /> -->  
    <ChildRec>  
        <InnerFieldElement1>Inner1</InnerFieldElement1>   
        <InnerFieldElement2>Inner2</InnerFieldElement1>  
    </ChildRec>  
    <FieldElem5 />  
</MyRec>

If this is passed to the flat file disassembler, the delimiters for FieldElem4 will not be preserved but subsequent records will be delimited as expected.

,,Val,,Inner1,Inner2,,

See Also