Node Design Guidelines
hbeeken edited this page Nov 3, 2014
·
6 revisions
There are three types of node.
- an Input node sits at the start of a flow. It is typically triggered by some external event occurring.
- a Function node sits in the middle of a flow. It performs actions based on the incoming messages and passes on the resulting messages. One sub-type of Function node is a Query node that is used to retrieve data from an external API, such as a database query or a web services call.
- an Output node sits at the end of a flow. It consumes the incoming messages and triggers an action.
(under development)
- data in the node settings overwrites that in an incoming message
- If data in the incoming message is not valid, the node should......? (Question: in this case should the node issue a warning? or an error? and/or change its status? or ignore this silently?)
Work in progress: