The visually larger Input and Output components represent service’s HTTP request and response. You should place your service logic in between the two components.

Input and Output are two special components of Data Service REST job. Their attribute editors allow alternative access to the REST endpoint input and output configuration eliminating the need to switch to the Endpoint Configuration tab.

The Input component can provide you with a binary stream of input data, or the whole request body as a string, or parse the payload as JSON into a variant data field.

If there is an edge connected to the parameters port, the Rest Job parameters are automatically parsed and converted to appropriate CTL2 types. Otherwise, conversion has to be done manually using the available conversion functions. See CTL2 data types and Conversion functions.

If the component reading the request body is in the lowest phase of the job, the same as the Input component, then the data can be read before the whole request body arrives. This is useful when the request body has a larger size and/or the network connection is slow. If the component reading the request body is not in the lowest phase of the job then the whole request body will be read and cached by the input component.

If you connect any edge to a port on the Output component, all records flowing through this edge will be used as a response payload. Before sending, the records are automatically serialized using one of the automatic serialization formats (JSON, XML or CSV) according to Endpoint Configuration.

For detailed information about data serialization, see the Generating response content section.