Connect+ uses two kinds of expressions: spelExpression for the transformation block and
NiFi expression for other fields.

Creating expression

The table below provides sample expressions along with additional information to assist you in understanding and creating new expressions as needed.

Arithmetic expressions

String operations

Conditional Operations

Complex expressions

Using SpEL expressions in data transformation

In the data transformation block, you can use expressions for the following:

• Filtering data
• Mapping attribution identifier type
• Mapping header that has consistent values

Filtering data

You can use filter expression to filter the data that needs to be processed.

Consider a transaction dataflow that includes a Financial status header with values such as paid, refunded, or not paid.

To exclusively process dataflow for transactions that are either paid or not paid, in the Transformation block, you can use the filter expression ${header_value:equals('paid'):or(${header_value:equals('not paid')})} . This expression should be entered in the Filter tab within the transformation text box.

Mapping attribution identifier type

Every transaction or customer addition is linked to a specific store or till. The details of these stores or tills are configured in the attribution block. Here, you have the flexibility to use expressions to define constant values for the identifier type.

For example, consider the scenario of a customer addition dataflow and the store code information is found under the 'SC_NUMBER' column in the source file. Here, you would set the identifierType as 'STORE_CODE' and map the identifierValue to the 'SC_NUMBER' column.

In this case, use the expression const{STORE_CODE} in the identifierType field and specify 'SC_NUMBER' as the identifierValue. This setup ensures that the attribution of 'STORE CODE' is consistently associated with the 'SC_NUMBER' column in your data.

Mapping header that has consistent values

Consider a scenario where you're adding transaction data using the "add transaction dataflow," and all the transactions share the same currency code, INR. The challenge here is that there's no specific header column in the source file indicating the currency code; instead, this information was conveyed via email.

In this situation, you can map the header to the API field by performing the following:

1. Add a new column to your source file with the header "Currency Code".
2. Populate this new column with the value "INR" for all transactions.
3. Map this "Transaction Type" header to the API field currencyCode.

This way, you explicitly specify that the currency code for all transactions is INR. However, there's a more efficient way to achieve this using an expression.

Simplify with expressions

You can streamline this process by using the expression const{INR} in the API field currencyCode. This expression communicates that the value for the API field currencyCode is consistently "INR" for all transaction data. This approach eliminates the need for additional columns and simplifies the mapping process.

Avoiding common errors in expression

The examples below can help you to avoid errors when working with the MVEL expressions.