AnsweredAssumed Answered

HTTP connection via Groovy custom script

Question asked by rkarunakaran925730 on Jan 3, 2018
Latest reply on Jan 5, 2018 by mmorthala189353


The problem:

In a sub-process, when implementing pagination with repeated HTTP calls in a loop, the page counter needs to be stored somewhere. It seems that it cannot be a document property cause it can be lost after HTTP call (see this and this). The workaround appears to be to use dynamic process property together with a flow control shape and that ensures that only one document is processed at a time. However as a result records will not be processed concurrently and this can add significant overhead 


I was thinking of creating a reusable custom script that calls a HTTP url (using HttpUrlConnection) taking as parameters the url, method, content type etc. and returning the HTTP response. This is similar to a connector but perhaps simpler to implement. The script will ensure that document properties are passed as without the need for flow control + dynamic process properties



What are the licensing implications since the above works with other shapes without the need for connector. The other option is to create a custom HTTP connector that can retain the document properties after HTTP call in sub-processes