Nanoflows - Retrieve by association - Strange behaviour

Hi all, While playing with nanoflows I stumbled across something very unexpected. If I do a retrieve over association for a non persistent object in a nanoflow a xas request "retrieve_by_path" is issued to the server. Furthermore, all the associated objects are sent to the server ??? Is this the expected behavior. If all the data is already available in the client why is it necessary to send it to and back from the server instead of getting it from the client state? Any ideas why this could be happening? Is this a bug? -Andrej   
2 answers

To respond to the answer by Joel,

If it is true that nanoflows are ment to be used in offline apps only then Mendix should stop trying to to sell them as microflows that run in your browser.  Just take a look at their official documentation - I will just quote one line

Nanoflows also offer great value in online applications (for example, for UI logic, validations, calculations, and navigation). 



Hi Andrej, Nanoflows are used for Offline apps. It's really powerful (we're building an offline app for one of our customers) but it has it's limitations/things you need to know.

  • Only persistent data can be used. So maybe the non-persistent entity that is is available is a bug in mendix.
  • Mendix calculates the entities (based on pages and nanoflows) that must be stored in de offline app and the synchronization makes sure all changed data is sent to the server and all authorised data is retrieved from the server.
  • Nanoflows are always running with Entity Access! A change on an attribute requires write rights for the user.

So using a nanoflow in an online page isn't really helping, I even think that's not ment to be.

The new Mendix app for iOS and Android contains the developer app functionality and supports offline apps. You know scanning the QR code from the modeler to test the app on your phone.