"Runnable" data packages

[psychemedia]

"Runnable" data packages

[pwalsh]

Hi @psychemedia

have you looked at https://github.com/frictionlessdata/datapackage-pipelines ?

If I understand your use case correctly, DPP could be used to do exactly what you want, with the following additions on top of the base library:

• A custom processor to open the current data package being streamed in R Studio
• A custom processor to open the current data package being streamed in a Jupiter Notebook
• Have a docker file that calls dpp with a pipeline spec, and takes a few env vars (path to the DP to open, open in R Studio or Jupyter, etc.)

We've love to see such a use case and even contributions of processors to do these things!

[psychemedia]

@pwalsh Not played with that, no... will take a look...

[psychemedia]

@pwalsh Actually, the datapackage isn't opened into Jupyter or RStudio environment directly. It's loaded into the database container, which is then linked to, and accessible from, the application environment(s).

That's why I posted to this repo...;-)

So the question more succinctly is: is there a simple, automated recipe for populating a MySQL or Postgres container with a datapackage so that someone who doesn't know how to set up a database can create a running containerised instance of a database containing the datapackage contents from a simple command?

[roll]

Thanks! It's MOVED to https://github.com/frictionlessdata/implementations/issues/18