Run configuration for snakemake files

[iromeo]

Run configuration for snakemake files

[iromeo]

Not sure that we really need this feature (i.e. run feature or rule from the context menu). If someone has any thoughts, you are welcome.

[majkelx]

I would love it

[iromeo]

Thank you for the feedback, but what for do you need it what is your use case? Is it educational or for real data processing?

[iromeo]

When I run real pipelines they took significant time (e.g. hours or days), so I don't want to keep my IDE opened all this time. Moreover, I often run them on the cluster so I don't such run configurations and I prefer launching snakemake pipelines from a command line.

In PyCharm (and other JetBrains IDEs) there is a terminal window in IDE, so it is quite convenient (for me) to launch the pipeline on test data in that window.

[majkelx]

I'm astronomer and using a lot of custom python code for data processing, mostly astronomic calibration/reduction pipelines.
I found snakemake to be great replacement of GNU make, especially we are trying to get rid of preparing shell tools for each task (use run: rather than shell: in snakemake language).
So running and debuging code executed from snakemake workflow in PyCharm would be very helpfull in the pipelines development.
Usually, I have some test data in the project, and run/test/debug the code from unit tests. But in the case of snakemake, the natural way to feed the pipeline code with data is via snakemake. Also some layer of code is now snakemake dependent.
I have seen #142 and #137, I understend that full-fledged debugger for snakemake isn't piece of cake, but... I would love it ;)

[iromeo]

@majkelx Ok, thx for the explanation

[majkelx]

Just for setting breakpoints in code called from snakemake and debugging it, I created dummy runner runsnakemake.py:

  from snakemake import main
  main()

And I'm using it in Run/Debug configuration to start snakemake under debuger.

This solves my only "real" problem.