Can I use "append instead of overwrite" on a filter recipe to store results from a Twitter stream?

[Alex_Combessie]

I run my filter recipe periodically. I would like to use append instead of overwrite to store only new tweets (not what is already stored with the recipe output). Is it possible?

[jrouquie]

Hi,

this looks like a use case for partitioning. Note this is an advanced concept of DSS.

Imagine one uses "append", discovers a bug, fixes it, runs the flow again: the output dataset would then contain both the buggy and the correct data.

[DvMg]

Do you have an example of how can do that?