Setting up a development enviroment to run LSST science pipeline code from your IDE: is this possible?

Dears,

I am trying to escape the fiendish jupyter browser-based IDE, and setup my favorite editor to run LSST pipeline code locally or on the cloud. In essence what I want to do is run the tutorial notebooks over PyCharm. I am unsure if at this point this is possible at all.

What I did:
Looking around I have found docker images for the lsst pipeline and instructions to set it up. If it is possible to set up this container to a workable local environment I’m well, since I can instruct my IDE to run in it. I experimented with the image a bit, installed lsst-rsp with no problem and so forth. However, at moment, I am unable to use both the butler (missing configuration dp02,) and the TAP service (no schema) from the container.

So here’s my questions:

  • Am I on the right track? Is it possible to run LSST code locally through docker? If so, is there any resource I’m missing on setting up a workable docker-based enviroment for experimenting with dp02 data?
  • If the answer to the point above is NO. has someone succeed in working with the RSP code and data from your local IDE? How did you got there?

Please pardon the possibly noobish question, am new here…

Bests,
Peppe

You can use the Docker images as a means of getting pre-packaged Science Pipelines code, but they are not complete in terms of providing access to data or any ancillary services.

In addition, we are moving away from the hub.docker.com hosted images to ghcr.io hosted images. Transitioning to ghcr as OCI repository

1 Like

Right, there’s also the matter of access to the DP0.2 data. For TAP, there are instructions for remote DP0.2 access:

https://dp0-2.lsst.io/data-access-analysis-tools/api-intro.html#use-of-pyvo-with-the-rsp-tap-service

Thank you for your replies! I also found this. I guess there really is no supported way to work with DP0.2 data from an IDE right now. That’s unfortunate but I’ll get over it :).

Bests,
Peppe