Index Functoid

The Index functoid enables you to select information from a specific record in a series of records. Each Index functoid collects information from a single field.

Certain records typically occur many times in an input file. For example, in a weather report, the DailySummary element might occur many times. The DailySummary element might include attributes for the temperature, the barometric pressure, and the wind speed. The following code is an example of a weather report.

<ns0:WeatherReport xmlns:ns0="http://IndexFunctoid.WeatherReport">  
    <DailySummary Pressure="80" Windspeed="10" Temperature="20" />  
    <DailySummary Pressure="78" Windspeed="20" Temperature="23" />  
    <DailySummary Pressure="77" Windspeed="16" Temperature="24" />  
</ns0:WeatherReport>  

In the underlying schema, the Max Occurs property for the DailySummary record would be set to unbounded to indicate a recurring or looping record. BizTalk Mapper compiles this record as a loop.

Suppose you want to collect weather information for the first two DailySummary records of the weather report. In BizTalk Mapper, each attribute from the DailySummary record of the incoming source schema can be connected to an Index functoid. In turn, each Index functoid can specify the DailySummary record from which to draw the information: the first or second. The Index functoids can then be connected to the appropriate fields of the destination schema.

The following figure shows Index functoids used in this way.

Index Functoid Example

To get the daily summary information for the first day, the upper set of three Index functoids have their index values set to 1. To get the daily summary information for the second day, the lower set of three Index functoids have their index values set to 2.

Index functoids use the Configure <Functoid> Functoid dialog box to set their input parameters. The first input parameter identifies a field within a looping record in the source schema. The second and succeeding input parameters specify the particular record. You can specify multiple index values to select a record within nested repeating structures. The index value for the innermost structure is the second parameter. The index value for the next outermost structure would be the third parameter, and so on. For example, suppose that the preceding DailySummary records were inside WeeklyData records. To retrieve the Pressure from the first DailySummary in the second WeeklyData, the second parameter would be 1 and the third parameter would be 2.

Notice that this example assumes the Pressure field does not repeat. If the field did repeat, the indices would be off—the count would begin with the Pressure field, rather than the Daily Summary.

Note

Although an index sequence input parameter is typically a constant, it is possible to use a link from a node in the source schema. If this link comes from a looping record that is not a parent of the first input parameter, the index sequence input value comes from the first instance of the node in the input instance message.

Note

The value of the index sequence input is always in relation to the current context in the source document.

Important

An Index functoid must have as many index values as there are parent nodes from the field level to the first level below the root node. For example, in the multiple weather report instance message, two index values are required. In the single weather report instance message, only one index value is required. Failure to set the required number of index values of an Index functoid creates output based on the first node in the source instance message that matches the first input parameter of the Index functoid.

See Also

How to Add Index Functoids to a Map
Advanced Functoids
Iteration Functoid
Record Count Functoid